CN117994016A - Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model - Google Patents
Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model Download PDFInfo
- Publication number
- CN117994016A CN117994016A CN202211324673.3A CN202211324673A CN117994016A CN 117994016 A CN117994016 A CN 117994016A CN 202211324673 A CN202211324673 A CN 202211324673A CN 117994016 A CN117994016 A CN 117994016A
- Authority
- CN
- China
- Prior art keywords
- credit
- past
- month
- months
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 274
- 238000013058 risk prediction model Methods 0.000 title claims abstract description 23
- 238000006243 chemical reaction Methods 0.000 claims abstract description 518
- 238000012216 screening Methods 0.000 claims abstract description 212
- 238000003066 decision tree Methods 0.000 claims abstract description 28
- 238000009795 derivation Methods 0.000 claims abstract description 4
- 125000004122 cyclic group Chemical group 0.000 claims description 91
- 238000004364 calculation method Methods 0.000 claims description 63
- 230000008569 process Effects 0.000 claims description 40
- 238000007477 logistic regression Methods 0.000 claims description 38
- 238000010276 construction Methods 0.000 claims description 35
- 230000006399 behavior Effects 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 26
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 24
- 239000010931 gold Substances 0.000 claims description 24
- 229910052737 gold Inorganic materials 0.000 claims description 24
- 230000009466 transformation Effects 0.000 claims description 23
- 230000009467 reduction Effects 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 210000001061 forehead Anatomy 0.000 claims description 9
- 238000011437 continuous method Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 230000003542 behavioural effect Effects 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 2
- 238000004513 sizing Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 97
- 230000008859 change Effects 0.000 description 24
- 238000012549 training Methods 0.000 description 23
- 238000011161 development Methods 0.000 description 22
- 238000012217 deletion Methods 0.000 description 20
- 230000037430 deletion Effects 0.000 description 20
- 230000000694 effects Effects 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 11
- 230000008901 benefit Effects 0.000 description 11
- 230000007423 decrease Effects 0.000 description 8
- 238000013461 design Methods 0.000 description 8
- 230000001965 increasing effect Effects 0.000 description 8
- 230000036961 partial effect Effects 0.000 description 8
- 230000002596 correlated effect Effects 0.000 description 7
- 238000002347 injection Methods 0.000 description 6
- 239000007924 injection Substances 0.000 description 6
- 230000007547 defect Effects 0.000 description 5
- 238000012850 discrimination method Methods 0.000 description 5
- 230000007774 longterm Effects 0.000 description 5
- 238000012790 confirmation Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000013077 scoring method Methods 0.000 description 3
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007717 exclusion Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The application relates to a retail credit Beta model and a method for predicting retail credit risk by using the model, which comprises the following steps: a data acquisition step of acquiring retail credit prediction data of a sample to be predicted; classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating credit violation probability; and a credit violation probability calculating step, wherein retail credit prediction data are substituted into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted. The application also relates to a method for constructing a retail credit risk prediction model, which comprises the following steps: the method comprises a data acquisition step, a data derivation step, a characteristic preliminary screening step, a preliminary screening data conversion step, a characteristic fine screening step and a credit default probability modeling step.
Description
Technical Field
The application relates to a credit risk management and control system and a credit risk management and control method, which can assist a financial institution to make more accurate risk decisions and accelerate the digital transformation process of the financial institution. In particular, the present application relates to a retail credit risk prediction model.
Background
In the current large environment of aggressive consumer credit development, the manual approval mechanisms of some financial institutions have failed to cope with the increasing credit demand, and thus it is highly desirable to promote the intelligent pneumatic control capability of financial institutions. Financial institutions wish to build a credit full-flow risk slow release mechanism from customer prescreening, pre-credit review, mid-credit approval, post-credit management to early credit-harvest stages.
If the scoring system can be developed based on the principles of early recognition, early warning, early discovery and early treatment, credit business can be monitored and managed quickly and conveniently, and the business volume, competitive advantage and asset quality of a financial institution can be improved on the basis of controllable risk.
However, the construction of the scoring system has extremely strong dependence on data and technical aspects, and the diversity and coverage of data dimensions, modeling skills and methodology directly influence the final stability and ordering of the scoring system. Some financial institutions accumulate less for intelligent wind control experience of retail business and have weaker wind control capability. In the actual application process, the management and control technical problems that the financial institutions cannot fully exert the internal data value, the accuracy and stability of the model cannot be effectively improved and the like due to factors such as the lack of data mining analysis capability, the weak risk modeling technology and the like exist. This is also one of the major obstacles facing small and medium-sized financial institutions in the direction of digital transformation.
Disclosure of Invention
In view of the above-described deficiencies of the prior art, the present application is directed to a credit risk management system and method that provides effective risk management for financial institutions. The credit risk prediction method and the credit risk prediction system are based on massive data of a large bank, and the risk law refined by utilizing a statistical principle has popularization significance.
The credit risk management and control method is suitable for credit risk management and control in credit business, and particularly aims at consumption credit business scenes. The consumer loan generally refers to a personal consumer loan, i.e., a rmb loan business issued by a bank to a personal customer for a specified consumer purpose, including consumer personal loans such as personal housing, automobiles, general learning-aid loans, and the like. The consumer lending business is considered to be a good property of the bank in the early stage of development because of low poor loan rate and high information rate. But during their development the risk of personal loans was found to be far greater than expected. The individual consumption credit objects are scattered, the individual loan amount is small, the business volume is large, the operation is complex, and the personal information system and the credit investigation system of China are not perfect, so that a bank has to put in great effort to know the credit investigation condition of the individual when carrying out the consumption credit business, or more complicated procedures and strict guarantee and mortgage conditions are set, thereby greatly improving the transaction cost of the bank.
To avoid the credit risk, a credit rating method may be used, for example, to objectively evaluate the credit rating of the borrower when issuing the credit business. Other popular scoring model construction processes in the current market are often limited by the disadvantages of small modeling sample number, single data source, high data dimension homogeneity and the like. Meanwhile, the risk assessment model on the market at present mostly uses data with weak financial properties, namely, models are built based on non-credit transaction data such as intelligent terminal equipment data, social platform data, online shopping mall data and non-overdue prediction targets, and a prediction result and an actual credit overdue situation often have larger deviation.
The method and system of the application are based on the development of highly stable, high coverage data samples, and systematically innovate the relatively mature credit management system at present, and predict the potential credit risk of financial institutions in the consumption credit business by using credit samples containing various business forms. Specifically, the application uses the consumer credit group sample modeling to mine the credit and asset characteristics of consumer credit customers (including credit cards, consumer credits and the like) and constructs a credit scoring method which is more suitable for financial institutions with consumer credits as the main business type.
Specifically, the present application may further predict the probability of future credit overdue with (1) a customer base for which no credit overdue has occurred in the applied credit business and the current credit usage is < 100% and the current monthly asset size is < 500 yuan; (2) A customer group who has applied for credit business and does not have credit overdue, wherein the current credit utilization rate is less than 100 percent, and the current monthly asset size is more than or equal to 500 yuan predicts the probability of the future occurrence of the credit overdue; (3) A customer group who has applied for credit business and does not have credit overdue, wherein the current credit utilization rate is more than or equal to 100 percent and the current month asset size is less than 500 yuan predicts the probability of the future occurrence of credit overdue; (4) The method comprises the steps that a customer group who does not generate credit overdue in an applied credit business, the current credit utilization rate is more than or equal to 100%, and the current monthly asset size is more than or equal to 500 yuan predicts the probability of generating credit overdue in the future; (5) A customer base who has applied for credit services and has developed a credit overdue predicts the probability of developing a credit overdue in the future; (6) The 6 credit violation probabilities, the probability that a customer group not applying for a credit service predicts that it will occur a credit overdue, were developed for the prediction objective (i.e., objective variable).
Compared with the existing models in the market, the method and the system of the application have great improvement in the aspects of distinguishing degree and stability of predicting the occurrence of the overdue credit of the specific guest group for 60 days and more.
The application relates to the following technical scheme:
1. A method of calculating retail credit risk, comprising:
A data acquisition step of acquiring retail credit prediction data of a sample to be predicted;
Classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating credit violation probability;
and a credit violation probability calculating step, wherein retail credit prediction data are substituted into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted, and the credit violation probability is preferably the credit violation probability.
2. The method of item 1, further comprising:
after calculating the credit violation probability, calculating a credit score of the sample to be predicted, for calibrating the calculated credit violation probability to a normalized score of 0-1000 points.
3. The method according to item 1 or 2, wherein,
The retail credit prediction data comprises original retail credit prediction data of a sample to be predicted and derived retail credit prediction data is processed based on the original retail credit prediction data;
preferably, the original retail credit prediction data includes:
credit card based data, which is based on all available data in the sample user's credit card creation process and use process,
Personal loan class base data, which is all available data based on the sample user's loan application case and usage behaviors,
Customer base information class base data, which is data based on the properties of the sample user itself, but not directly related to behavior at the financial institution,
Personal financial property type base data, which is sample user data of all other financial properties and financial transaction types not related to credit cards and loans at a financial institution.
4. The method according to any one of items 1 to 3, wherein,
Processing derived retail credit prediction data based on the original retail credit prediction data refers to data obtained by processing the collected original retail credit prediction data based on a time dimension, a space dimension, a frequency dimension, and a statistical information dimension;
preferably, the derived retail credit prediction data includes, but is not limited to:
derived retail credit prediction data processed based on sample relationship lengths,
Derived retail credit forecast data processed based on time interval class variables,
Derived retail credit forecast data processed based on the degree of sample behavioral frequency,
Derived retail credit prediction data processed based on the sample current point in time conditions,
Derived retail credit forecast data processed based on sample duration behavior,
Derived retail credit prediction data processed from the sample data based on the statistical information dimension.
5. The method according to any one of items 1 to 4, wherein,
The retail credit prediction data is selected from one or two or three or four or five or six or seven of the following:
The current time point credit account balance, the maximum expiration date of the past 12 month credit customer, the time point balance loop ratio of the past 6 month credit account, the maximum continuous reduction month rate, the current credit card remaining available amount, the past 12 month wage maximum, the cyclic credit past 6 month line usage minimum, the past 12 month line usage maximum continuous month rate >70%, the past 3 month monthly credit current month refund amount/month average asset size value, the cyclic credit past 12 month line usage >10% month rate, the credit past 12 month line usage >70% month number, the past 6 month average asset size/past 12 month average asset size, the past 3 month line usage >90% month number, the past 3 month credit account time point minimum balance, the cyclic credit account usage > 3% month average asset size/past 12 month average asset size value average property size of past 12 months, minimum balance of past 3 month financial account to date month, minimum balance of past 3 months monthly credit due month payoff/month average property size, past 6 month deposit account time point minimum balance, whether to send out wages, past 6 month financial product balance/past 12 month financial product balance, month number of cyclic credit past 6 month line usage >70%, past 3 month credit customer overdue balance average, percentage of past 6 month credit customer overdue balance to loan amount, credit customer current account age, credit past 3 month repayment amount maximum value, past 6 month gold account holding month number, past 12 month investment financial account average balance, past 3 month asset size/past 12 month asset size average.
6. The method according to any one of items 1 to 5, wherein,
The step of classifying the samples to be predicted comprises the sub-steps of:
Whether the sample to be tested is a customer who has applied for credit services at a financial institution;
Whether the sample to be tested is a customer who has a credit overdue;
whether the current month credit utilization rate of the sample to be tested is more than or equal to a percent;
whether the current month asset scale of the sample to be tested is more than or equal to b;
Classifying samples to be predicted based on the substeps to determine a submodel for calculating credit violation probability, wherein the sequence of the substeps can be arbitrarily set on the premise of ensuring reasonable business logic;
The samples to be predicted are preferably classified in the following order:
firstly, judging whether the sample to be tested is a customer who applies for credit business at a financial institution;
Then judging whether the sample to be tested is a customer with overdue credit;
then judging whether the current month credit utilization rate of the sample to be tested is more than or equal to a percent;
And then judging whether the current month asset size of the sample to be tested is more than or equal to b.
7. The method according to any one of items 1 to 6, wherein,
And substituting the feature conversion of the retail credit prediction data into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted, wherein the feature conversion step comprises the following steps:
And selecting a WOE mode, a dummy feature mode or a continuous mode to perform feature conversion based on the feature type of the retail credit prediction data which is required to be substituted into the credit violation probability submodel.
8. The method according to item 7, wherein,
The feature conversion in the continuous mode comprises the following steps: the continuous feature conversion is carried out in a mode of directly selecting an original value, calculating the square of original data, calculating the square root of the original data, calculating the cube root of the original data or calculating the natural logarithm of the original data.
9. The method according to any one of items 1 to 8, wherein,
The credit violation probability submodel is a model constructed based on existing user populations using logistic regression based on sample retail credit prediction data and credit violation probabilities.
10. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the current time point credit account balance, the maximum overdue number of the past 12 month credit client, the time point balance loop ratio of the past 6 month credit account, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the past 12 month wage maximum value, the cyclic credit past 6 month amount use ratio minimum value of one, two, three, four, five or six.
11. The method of item 10, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on the current time point deposit account balance, the maximum overdue number of the past 12 month credit clients, the time point balance loop ratio of the past 6 month deposit account, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the maximum wage value of the past 12 months and the minimum cyclic credit past 6 month amount use rate,
Preferably, converting the balance of the deposit account at the current time point in a continuous mode; converting the maximum overdue number of the past 12-month credit customer in a continuous mode; converting the point balance ring ratio of the past 6 month deposit account with the maximum continuous reduction month number ratio in a continuous mode; converting the residual usable credit of the current credit card in a continuous mode; converting the maximum wage value of the past 12 months in a dummy feature mode; converting the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit in a continuous mode;
Further preferably, the method comprises converting the current time point deposit account balance into a calculation method using natural logarithm of the current time point deposit account balance by adopting a continuous mode; converting the maximum overdue number of the past 12-month credit customer into a calculation mode using an original value for the maximum overdue number of the past 12-month credit customer in a continuous mode; converting the time point balance ring ratio of the past 6 month deposit account into a calculation mode of taking a cube root by adopting a continuous mode according to the time point balance ring ratio of the past 6 month deposit account; converting the current credit card residual usable amount into a calculation mode of taking a cube root for the current credit card residual usable amount; the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit is converted into a calculation mode of taking the square root of the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit by adopting a continuous mode.
12. The method according to item 11, wherein,
Substituting the values converted by the six characteristics of the current time point deposit account balance, the maximum overdue number of the past 12 month credit clients, the time point balance loop ratio of the past 6 month deposit accounts, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the wage maximum value of the past 12 months and the cyclic credit past 6 month amount use ratio minimum value into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
13. The method of item 12, wherein,
Wherein the submodel is shown in formula 1 below:
where k is the number of features of the model entered, preferably k is 6,
Alpha is an intercept term, preferably in the range of values (0.20792, -0.57608), most preferably-0.18408;
β1 is the corresponding coefficient of the balance of the deposit account at the current time point, and the preferred value range is (-0.41409, -0.45329), and the most preferred value range is-0.43369;
β2 is the maximum expiration number corresponding to the last 12 months of credit customers, preferably ranging in value from (-0.65896, -0.69816), most preferably-0.67856;
β3 is the corresponding coefficient of the point balance ring ratio of the past 6 month deposit account with the maximum continuous reduction month number, and the preferable value range is (-0.47066, -0.50986), and the most preferable value range is-0.49026;
Beta 4 is the corresponding coefficient of the residual usable credit of the current credit card, and the preferred value range is (-0.15071, -0.18991), and most preferably-0.17031;
beta 5 is the maximum payroll corresponding coefficient for the past 12 months, the preferred range of values is (-1.08474, -1.16314), and most preferably-1.12394;
Beta 6 is the corresponding coefficient of the minimum value of the limit use rate of the cyclic credit for the past 6 months, the value range is (-0.51313, -0.70913), and the most optimal value is-0.61113;
x1 is a natural logarithmic conversion value of the balance of the deposit account at the current time point generated in the feature conversion step;
x2 is the original value of the maximum overdue number of the past 12 month credit customer generated by the feature conversion step;
x3 is a cube root conversion value of the time point balance ring ratio of the past 6 month deposit account generated in the feature conversion step, which is the maximum continuous reduction month number ratio;
x4 is the cube root conversion value of the residual usable credit of the current credit card generated in the feature conversion step;
x5 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step;
x6 is the square root transformed value of the minimum usage of the credit for the last 6 months of the cycle generated by the feature transformation step.
14. The method according to item 13, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
15. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: one, two, three, four or five of a maximum number of consecutive months of past 12 month line usage >70%, a minimum of past 6 month line usage of recurring credits, a current credit card remaining available line, a past 3 month monthly credit this month should be refunded/month average asset size value, a current point-in-time deposit account balance.
16. The method of item 15, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on the maximum continuous month number of which the use rate of the credit line of the past 12 months is more than 70%, the use rate of the credit line of the past 6 months of the cyclic credit is the minimum value, the current credit card remaining usable amount, the amount to be paid per month of the past 3 months monthly credit, the asset scale value of the current time point deposit account balance,
Preferably, converting the maximum continuous month number of which the use rate of the forehead of the past 12 months is more than 70% in a continuous mode; converting the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit in a continuous mode; converting the residual usable credit of the current credit card in a continuous mode; converting the value of the amount to be refund per month of the past 3 months monthly credits per month of the average asset scale in a continuous mode; converting the balance of the deposit account at the current time point in a continuous mode;
Further preferably, the conversion of the maximum continuous month number of the past 12 month line usage >70% is a calculation of the maximum continuous month number usage of the past 12 month line usage >70% by using the original value, the conversion of the minimum cyclic credit line usage of the past 6 months is a calculation of the minimum cyclic credit line usage of the past 6 months by using the square root, the conversion of the current credit card remaining usable line by using the continuous method is a calculation of the current credit card remaining usable line by using the natural logarithm, the conversion of the past 3 month monthly credit present month account amount/month average asset scale value by using the cube root for the past 3 month monthly credit present month account amount/month average asset scale value, and the conversion of the current time point account by using the continuous method is a calculation of the current time point account use natural logarithm.
17. The method according to item 15, wherein,
The five feature-converted values of the maximum continuous month number of which the use rate of the credit of the past 12 months is more than 70%, the use rate of the credit of the cycle is the minimum value of the credit of the past 6 months, the current credit card remaining usable amount, the amount of money to be paid per month of the past 3 months monthly credit, the asset scale value of the deposit account balance of the current time point are substituted into a submodel constructed by adopting logistic regression based on sample retail credit prediction data to calculate the default probability of the sample to be predicted.
18. The method according to item 17, wherein,
The submodel is shown in the following formula 2:
Where k is the number of features of the model entered, preferably k is 5;
Alpha is the intercept term, preferably in the range (-0.27492, -0.66692), most preferably-0.47092;
β1 is the maximum continuous month number corresponding coefficient with the use rate of the forehead of the past 12 months >70%, and the preferred numerical range is (0.19904-0.03616), and the most preferred numerical range is 0.081443;
β2 is the minimum corresponding coefficient of the usage rate of the line of the past 6 months of the cyclic credit, and the preferred value range is (0.106, -0.09), and the most preferred value range is 0.008002;
beta 3 is the corresponding coefficient of the residual usable credit of the current credit card, and the preferred value range is (-0.49447, -0.76887), and most preferably-0.63167;
β4 is a corresponding coefficient of the amount of credit payoff per month of the last 3 months monthly/month average asset size value, preferably in the range of values (0.24108, -0.22932), most preferably 0.005877;
beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the preferred value range is (-0.6779, -0.9131), and the most preferred value range is-0.7955;
x1 is the original value of the maximum consecutive months of which the rate of use of the line of the past 12 months generated in the feature conversion step is > 70%;
x2 is the square root conversion value of the minimum value of the usage rate of the line of the past 6 months of the cyclic credit generated by the feature conversion step;
x3 is the natural logarithmic conversion value of the current credit card remaining usable credit generated in the feature conversion step;
x4 is the cube root conversion value of the asset scale of the amount to be refund/month average for the last 3 months monthly credits generated by the feature conversion step;
x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step.
19. The method of item 18, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
20. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the minimum value of the usage rate of the credit for the line of the past 6 months, the usage rate of the credit for the line of the past 12 months of the cyclic credit is more than 10 percent, the usage rate of the credit for the line of the past 12 months is more than 70 percent, the asset size of the average asset of the past 6 months/the asset size of the average asset of the past 12 months, the usage rate of the line of the past 3 months is more than 90 percent, and one, two, three, four, five or six of the minimum balances of the time points of the deposit account of the past 3 months.
21. The method of item 20, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on a minimum value of the utilization rate of the line of the past 6 months of the cyclic credit, a month number proportion of the utilization rate of the line of the past 12 months of the cyclic credit of >10%, a month number of the utilization rate of the line of the past 12 months of the credit of >70%, an average asset scale of the past 6 months/an average asset scale of the past 12 months, a month number of the utilization rate of the line of the past 3 months of >90%, and a minimum balance of the time points of the deposit account of the past 3 months,
Preferably, the minimum value of the use rate of the line of the past 6 months of the cyclic credit is converted in a continuous mode; converting the month number occupation ratio of the cyclic credit with the use ratio of more than 10% in the past 12 months by adopting a continuous mode; converting the number of months with the credit limit use rate of more than 70% in the past 12 months in a continuous mode; converting the asset size of the past 6 months/the asset size of the past 12 months in a continuous mode; the method comprises the steps of converting the number of months with the use rate of the line of 3 months more than 90% in the past by adopting a dummy feature mode; converting the minimum balance of the past 3 month deposit account time point in a continuous mode;
Further preferably, the minimum value of the usage rate of the line of the past 6 months of the cyclic credit is converted into a calculation mode of taking a cube root for the minimum value of the usage rate of the line of the past 6 months of the cyclic credit, the month number proportion of the line of the past 12 months of the cyclic credit with the usage rate of >10% is converted into a calculation mode of taking an original value for the month number of the line of the past 12 months of the cyclic credit with the usage rate of >10%, the method comprises the steps of converting the month number of which the credit use rate is more than 70% in the past 12 months into a calculation mode of using a raw value for the month number of which the credit use rate is more than 70% in the past 12 months, converting the past 6 month average asset scale/the past 12 month average asset scale into a calculation mode of using a cube root for the past 6 month average asset scale/the past 12 month average asset scale, and converting the minimum balance of the past 3 month deposit account time points into a calculation mode of using a natural logarithm for the past 3 month deposit account time points by using a continuous mode.
22. The method according to item 20, wherein,
Substituting the values converted by six characteristics of the minimum value of the utilization rate of the line of the circulation credit for the past 6 months, the utilization rate of the line of the circulation credit for the past 12 months >10% of the month number proportion, the utilization rate of the line of the credit for the past 12 months >70% of the month number proportion, the average asset scale of the past 6 months/the average asset scale of the past 12 months, the utilization rate of the line of the past 3 months >90% of the month number proportion and the minimum balance of the time point of the deposit account of the past 3 months into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
23. The method of item 22, wherein,
The submodel is shown in equation 3 below:
Where k is the number of features of the model entered, preferably k is 6;
Alpha is the intercept term, preferably in the range (-0.23953, -0.27873), most preferably-0.25913;
; β1 is the minimum value corresponding coefficient of the rate of use of the line of the last 6 months of the cyclic credit, and the preferred value range is (-0.4522, -0.4914), and most preferably-0.4718;
β2 is a month number corresponding coefficient with a cyclic credit usage rate of >10% for the past 12 months, and the preferred value range is (-0.22933, -0.26853), and most preferably-0.24893;
Beta 3 is a month number corresponding coefficient with a credit usage rate of 12 months in the past of >70%, and the preferred value range is (-0.51155, -0.58995), and most preferably-0.55075;
Beta 4 is the corresponding coefficient of the average asset size of the past 6 months/the average asset size of the past 12 months, and the preferable numerical range is (-0.19469, -0.27309), and the most preferable numerical range is-0.23389;
beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, and the preferable numerical range is (-0.22323, -0.26243), and the most preferable numerical range is-0.24283;
β6 is the minimum balance corresponding coefficient at the time of the deposit account of the past 3 months, and the preferred value range is (-0.80785, -0.84705), and most preferably-0.82745;
x1 is a cube root conversion value of the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit generated in the characteristic conversion step;
x2 is the original value of the month number occupation ratio of the cyclic credit generated in the feature conversion step, wherein the use rate of the line of the past 12 months is more than 10%;
x3 is the original value of the number of months for which the credit generated in the feature conversion step has a usage rate of >70% for the past 12 months;
x4 is the cube root conversion value of the last 6 month average asset size/the last 12 month average asset size generated in the feature conversion step;
x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step;
x6 is the natural log-transformed value of the minimum balance of the past 3 month deposit account timepoints generated by the feature transformation step.
24. The method of item 23, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
25. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: average asset size over the past 12 months, minimum balance of past 3 month financial account from present month, minimum balance of past 3 month monthly credit due to present month payoff/average asset size, minimum balance of past 6 month deposit account time point, month number of past 3 month line usage >90%, whether to send out wages, one, two, three, four, five, six and seven of past 6 month financial product balances/past 12 month financial product balances.
26. The method of item 25, wherein the credit violation probability calculating step comprises:
The average asset scale of the past 12 months, the minimum balance of the past 3 months financial account, the minimum balance of the past 3 months monthly credit and the corresponding repayment/average asset scale of the month, the minimum balance of the past 6 months deposit account time point, the month number of the past 3 months use rate of >90 percent, whether to send out wages, the balance of the past 6 months financial products/the balance of the past 12 months financial products are subjected to characteristic conversion,
Preferably, the average asset size over the past 12 months is converted in a continuous manner; converting the minimum balance of the financial account of the past 3 months from the current month by adopting a continuous mode; converting the repayment/month-average asset scale of the monthly credit of the past 3 months by adopting a continuous mode; converting the minimum balance of the past 6 month deposit account time point in a continuous mode; the method comprises the steps of converting the number of months with the use rate of the line of 3 months more than 90% in the past by adopting a dummy feature mode; converting whether the generation wages adopt a dummy feature mode; converting the balance of the financial product of the past 6 months/the balance of the financial product of the past 12 months in a continuous mode;
further preferably, the method of calculating the natural logarithm of the average asset scale of the past 12 months is converted from the continuous method of calculating the natural logarithm of the average asset scale of the past 12 months, the method of calculating the natural logarithm of the minimum balance of the past 3 month financial account from the present month is converted from the minimum balance of the past 3 month financial account from the present month to the original value of the minimum balance of the past 3 month financial account, the method of calculating the square root of the average asset scale of the past 3 month monthly credit to the average asset scale of the past 3 month monthly credit to the average asset scale of the past 3 month, the method of calculating the minimum balance of the past 6 month deposit account from the minimum balance of the past 6 month deposit account, and the method of calculating the square root of the past 6 month financial balance/the past 12 month financial balance.
27. The method according to item 25, wherein,
Substituting values converted from seven characteristics of average asset size of the past 12 months, minimum balance of the past 3 month financial account, minimum balance of the past 3 month monthly credit due to repayment/average asset size of the month, minimum balance of the past 6 month deposit account time point, use rate of the past 3 month line >90%, whether to send wages, past 6 month financial product balance/past 12 month financial product balance into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
28. The method of item 27, wherein,
The submodel is shown in equation 4 below:
where k is the number of features of the model entered, preferably k is 7;
alpha is the intercept term, preferably the range of values is (3.5122,3.4338), most preferably 3.473;
β1 is the average asset size corresponding coefficient over the past 12 months, with a preferred range of values (-0.1731, -0.2123), most preferably-0.1927;
β2 is the corresponding coefficient of the minimum balance of the financial account of the past 3 months to the number of months, and the preferred value range is (-0.21993, -0.29833), and the most preferred value range is-0.25913;
β3 is the corresponding coefficient of the size of the assets that should be repaid/month-average for the past 3 months monthly credit, the preferred range of values is (-0.22933, -0.26853), and most preferably-0.24893;
β4 is the minimum balance corresponding coefficient at the time of the deposit account of the past 6 months, and the preferred value range is (-0.53115, -0.57035), and most preferably-0.55075;
Beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, and the preferable numerical range is (-0.27928, -0.35768), and the most preferable numerical range is-0.31848;
β6 is a payroll corresponding coefficient, preferably in the range (-0.21913, -0.25833), most preferably-0.23873;
beta 7 is the corresponding coefficient of the balance of the financial product in the past 6 months/the balance of the financial product in the past 12 months, and the preferable value range is (-0.55257, -0.63097), and the most preferable value range is-0.59177;
x1 is the natural logarithmic conversion value of the average asset size over the past 12 months generated by the feature conversion step;
x2 is the original value of the minimum balance of the financial account of the past 3 months generated in the feature conversion step from the number of months;
x3 is the cube root conversion value of the monthly payment/monthly average asset size of the past 3 months monthly credit generated in the feature conversion step;
x4 is a natural logarithmic conversion value of the minimum balance of the past 6 month deposit account time points generated in the feature conversion step;
x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step;
x6 is a dummy feature conversion value of whether to replace wages generated in the feature conversion step;
x7 is the square root conversion value of the past 6 month financial product balance/the past 12 month financial product balance generated by the feature conversion step.
29. The method of item 28, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
/>
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
30. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the cycle credit has a past 6 month line usage of >70% for one, two, three, four or five of the month number, past 3 month credit customer overdue balance average, past 6 month credit customer overdue balance percentage of the loan amount, credit customer current account age, credit past 3 month repayment amount maximum.
31. The method of item 30, wherein the credit violation probability calculating step comprises:
performing characteristic conversion on the month number of which the cycle credit has the past 6 month limit use rate of more than 70 percent, the past 3 month credit customer overdue balance average value, the percentage of the past 6 month credit customer overdue balance accounting for the loan amount, the current account age of the credit customer and the maximum value of the past 3 month repayment amount of the credit,
Preferably, the month number of which the use rate of the line of the past 6 months of the cyclic credit is more than 70% is converted in a continuous mode; converting the overdue balance mean value of the past 3-month credit client in a continuous mode; converting the percentage of overdue balances of past 6-month credit clients in the loan amount by adopting a dummy feature mode; converting the current account age of the credit client in a WOE mode; converting the maximum value of the repayment amount of the past 3 months of credit in a continuous mode;
It is further preferable that the method of continuously converting the number of months of the past 6 month line usage rate >70% of the cyclic credit into the method of calculating the use of the original value of the number of months of the past 6 month line usage rate >70% of the cyclic credit, the method of continuously converting the average value of the past 3 month credit customer overdue balance into the method of calculating the use of the natural logarithm of the average value of the past 3 month credit customer overdue balance, and the method of continuously converting the maximum value of the 3 month refund amount of the past 3 month credit into the method of calculating the use of the square root of the maximum value of the 3 month refund amount of the credit.
32. The method of item 30, wherein,
Substituting the numerical values converted by five characteristics of the past 6 month limit utilization rate of cyclic credit of 70 percent, the past 3 month credit customer overdue balance average value, the past 6 month credit customer overdue balance percentage of the loan amount, the current account age of the credit customer and the past 3 month repayment amount maximum value into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
33. The method of item 32, wherein,
The submodel is shown in equation 5 below:
Where k is the number of features of the model entered, preferably k is 5;
alpha is the intercept term, preferably the range of values is (2.9266,2.7698), most preferably 2.8482;
β1 is a month number corresponding coefficient with a cyclic credit usage rate of >70% for the past 6 months, preferably in the range of (0.05201, -0.06559), most preferably-0.00679;
β2 is the average corresponding coefficient of the overdue balances of the credit customers over the past 3 months, and the preferred value range is (0.03734, -0.04106), and most preferably-0.00186;
β3 is a corresponding coefficient of the percentage of the past 6 months credit customer's overdue balance to the loan amount, preferably ranging in value from (-0.10043, -0.21803), optimally-0.15923;
Beta 4 is the current account age corresponding coefficient of the credit client, the preferred numerical range is (0.03485, -0.20035), and the most preferred numerical range is-0.08275;
beta 5 is the maximum value corresponding coefficient of the repayment amount of the past 3 months of credit, and the preferred value range is (-0.36818, -0.44658), and most preferably-0.40738;
x1 is the original value of the number of months that the cyclic credit generated in the feature conversion step has a usage rate of the line of the past 6 months of > 70%;
x2 is a natural logarithmic conversion value of the average value of the overdue balances of the past 3-month credit customers generated in the feature conversion step;
x3 is a dummy feature conversion value of the percentage of the overdue balance of the past 6 month credit customer to the loan amount generated in the feature conversion step;
x4 is the WOE conversion value of the current account age of the credit client generated in the feature conversion step;
x5 is the square root conversion value of the maximum value of the repayment amount of the credit for the past 3 months generated by the feature conversion step.
34. The method of item 33, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
35. The method according to any one of the claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: one, two, three, four or five of the past 6 month gold account holding month number, the past 12 month investment financial account average balance, the past 12 month wage maximum, the past 3 month asset size/past 12 month asset size average, the current time point deposit account balance.
36. The method of item 35, wherein the credit violation probability calculating step comprises:
performing characteristic conversion on the number of holding months of the past 6 month gold account, the average balance of the past 12 month investment financial account, the maximum wage value of the past 12 months, the average value of the past 3 month asset scale/the past 12 month asset scale and the balance of the deposit account at the current time point,
Preferably, the number of the past 6 months of the gold account holding month is converted in a continuous mode; converting the average balance of the investment financial account of the past 12 months in a continuous mode; converting the maximum wage value of the past 12 months in a dummy feature mode; converting the average value of the past 3 month asset scale/the past 12 month asset scale in a continuous mode; converting the balance of the deposit account at the current time point in a continuous mode;
Further preferably, the calculation method is a calculation method in which the past 6 months of the number of gold account holding months is converted into the original value used for the past 6 months of gold account holding months by a continuous method; the average balance of the investment financial account of the past 12 months is converted into a calculation mode of taking natural logarithm by adopting a continuous mode; converting the past 3 month asset scale/past 12 month asset scale average value into a calculation mode using a cube root for the past 3 month asset scale/past 12 month asset scale average value by adopting a continuous mode; the current time point deposit account balance is converted into a calculation mode of using natural logarithm for the current time point deposit account balance by adopting a continuous mode.
37. The method according to item 35, wherein,
Substituting the five feature-converted values of the past 6 month gold account holding month number, the past 12 month investment financial account average balance, the past 12 month wage maximum value, the past 3 month asset scale/past 12 month asset scale average value and the current time point deposit account balance into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
38. The method of item 37, wherein,
The submodel is shown in equation 6 below:
Where k is the number of features of the model entered, preferably k is 5;
Alpha is the intercept term, preferably in the range (-3.0574, -3.2926), most preferably-3.175;
β1 is the corresponding coefficient for the number of months the gold account has over the last 6 months, the preferred range of values is (-0.76852, -0.96452), and most preferably-0.86652;
β2 is the average balance corresponding coefficient of the investment financial account of the past 12 months, the preferred value range is (-0.37245, -0.64685), and the most preferred value range is-0.50965;
β3 is the maximum payroll corresponding coefficient for the past 12 months, with a preferred range of values (-0.81714, -1.28754), and most preferably-1.05234;
beta 4 is the average corresponding coefficient of the past 3 month asset scale/the past 12 month asset scale, the preferred value range is (-0.0053, -0.2405), and most preferably-0.1229;
beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the preferred value range is (-0.92484, -1.04244), and the most preferred value range is-0.98364;
x1 is the original value of the number of months held by the gold account for the last 6 months generated in the feature conversion step;
x2 is a natural logarithm conversion value of the average balance of the investment financial account of the past 12 months generated in the characteristic conversion step;
x3 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step;
x4 is the cube root transformed value of the past 3 month asset size/past 12 month asset size average generated by the feature transformation step;
x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step.
39. The method of item 38, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
40. An apparatus for calculating retail credit risk, comprising:
The data acquisition module is used for acquiring retail credit prediction data of a sample to be predicted;
A module for classifying the sample to be predicted, for classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating a credit violation probability;
A credit breach probability calculation module for substituting retail credit prediction data into a credit breach probability submodel to calculate a credit breach probability for the sample to be predicted.
41. The apparatus of item 40, wherein the apparatus performs the steps of the method of calculating retail credit risk of any one of items 1 to 39.
42. A system for calculating retail credit risk, the system for calculating retail credit risk comprising: a memory, a processor, and a program stored on the memory and executable on the processor for the method of calculating retail credit risk, the program for calculating retail credit risk when executed by the processor implementing the steps of the method of calculating retail credit risk as recited in any one of items 1 to 39.
43. A computer storage medium having stored thereon a program for a method of calculating retail credit risk, the program for a method of calculating retail credit risk when executed by a processor implementing the steps of the method of calculating retail credit risk as recited in any one of items 1 to 39.
44. A method of constructing a retail credit risk prediction model for a consumer credit business, comprising:
a data acquisition step of acquiring raw retail credit prediction data of a sample used for constructing a model;
a data deriving step of processing derived retail credit prediction data based on the original retail credit prediction data;
A feature preliminary screening step of preliminarily screening all categories, i.e., all features, including the original retail credit prediction data and the derived retail credit prediction data to obtain preliminarily screened features;
a preliminary screening data conversion step of judging a conversion mode of the preliminarily screened features to confirm that one of a WOE conversion mode, a dummy feature conversion mode and a continuous conversion mode is adopted to perform feature conversion, and performing feature conversion by adopting a judged optimal mode for each preliminarily screened feature;
A feature fine screening step, namely performing deep screening on the features subjected to the primary screening after feature conversion to obtain the features subjected to the fine screening;
A credit violation probability modeling step, namely selecting a logistic regression mode for the combination of the characteristics after fine screening and the probability relation between the credit violations to carry out model construction, and confirming a mode for calculating the credit violation probability;
Wherein the data sample obtained in the data collection step is a customer sample that has used the consumer credit service.
45. The method of item 44, wherein,
In the data acquisition step, the raw retail credit prediction data of the sample for constructing the model obtained includes:
credit card based data, which is based on all available data in the sample user's credit card creation process and use process,
Personal loan class base data, which is all available data based on the sample user's loan application case and usage behaviors,
Customer base information class base data, which is data based on the properties of the sample user itself, but not directly related to behavior at the financial institution,
Personal financial property type base data, which is sample user data of all other financial properties and financial transaction types not related to credit cards and loans at a financial institution.
46. The method of item 44, wherein,
In the data deriving step, processing derived retail credit prediction data based on the original retail credit prediction data refers to data obtained by processing the collected original retail credit prediction data based on a time dimension, a space dimension, a frequency dimension, and a statistical information dimension;
preferably, the derived retail credit prediction data includes, but is not limited to:
derived retail credit prediction data processed based on sample relationship lengths,
Derived retail credit forecast data processed based on time interval class variables,
Derived retail credit forecast data processed based on the degree of sample behavioral frequency,
Derived retail credit prediction data processed based on the sample current point in time conditions,
Derived retail credit forecast data processed based on sample duration behavior,
Derived retail credit prediction data processed from the sample data based on the statistical information dimension.
47. The method according to any one of items 44 to 46, wherein,
The feature primary screening step comprises the following steps:
A first preliminary screening step of screening the features based on the data missing for each feature of the sample used to construct the model,
A second preliminary screening step of screening the features based on the condition that the single value of a certain feature sample is too high,
A third preliminary screening step, namely calculating the information IV value of each feature to carry out preliminary screening on the features;
The order of the first preliminary screening step, the second preliminary screening step and the third preliminary screening step may be any order,
A fourth preliminary screening step, wherein the features subjected to the first to third preliminary screening are subjected to preliminary screening by adopting a gradual discrimination algorithm;
And a fifth preliminary screening step, wherein the features after the fourth preliminary screening step are subjected to preliminary screening based on the coincidence condition of the risk characteristics of the features and the actual and real results of the samples for model construction.
48. The method of any one of items 44 to 47, further comprising:
A sample selection step for screening all users and acquiring a sample using the consumer credit service for model construction prior to the data acquisition step,
Preferably, the sample selection step includes classifying all users of the sample based on a decision tree, the classification basis including but not limited to:
Whether a user is a customer who has applied for registering credit services at a financial institution;
whether a certain user is a customer who has a credit overdue;
whether the current monthly credit utilization rate of a certain user is more than or equal to a percent;
whether the current monthly asset size of a certain user is larger than or equal to b.
49. The method according to any one of items 44 to 48, wherein in the preliminary screening data conversion step, the determination of the conversion pattern of the preliminarily screened features is made based on the concentration and the data type of the preliminarily screened features.
50. The method of item 49, wherein,
The primary screening data conversion step based on the judgment of the concentration and the data type comprises the following steps:
classifying the data type for each feature classifies each feature into character-type variables and numerical-type variables,
The character type variable is subjected to preliminary screening data conversion by adopting a dummy feature conversion mode,
The process of further classifying the numerical variables comprises the following sub-steps:
If the numerical variable has less than n values, performing primary screening data conversion by adopting a WOE conversion mode,
If the value of the numerical variable is more than n, further judging that if the numerical variable is converted into continuous variable with more values and the concentration degree of single value is more than m%, adopting a WOE conversion mode, if the concentration degree of single value is less than or equal to m%, adopting a continuous conversion mode,
Preferably, n and m are both positive integers, where n=5 to 10 and m=90 to 99.
51. The method of item 50, further comprising:
An optimal conversion method is selected for confirming the feature adopting the continuous conversion mode based on the correlation of the feature with credit violations under different continuous conversion modes to perform continuous feature conversion of the feature,
The continuous feature transformation is preferably performed by directly selecting the original value, calculating the square of the original data, calculating the square root of the original data, calculating the cube root of the original data, or calculating the natural logarithm of the original data.
52. The method of any one of claims 44 to 51, wherein the feature fine screening step comprises:
A first fine screening step, based on a stepwise regression algorithm, of screening features based on the significance of the features by the F test and the T test,
A second fine screening step of calculating a variance expansion factor based on each feature and eliminating features having a higher variance expansion factor to screen the features,
And a third fine screening step of analyzing whether the characteristic coefficient accords with the trend of the prediction result aiming at credit violations or not based on the logistic regression pair after the first fine screening step and the second fine screening step so as to further perform characteristic screening.
53. The method of any one of items 44 to 52, wherein the credit violation probability modeling step substitutes the features filtered by the feature fine screening step into a Sigmoid function for logistic regression to calculate a model of the credit violation probability.
54. An apparatus for constructing a retail credit risk prediction model for a consumer credit service, the apparatus comprising:
A data acquisition module for acquiring raw retail credit prediction data for a sample used to build a model;
a data derivation module for processing derived retail credit prediction data based on the original retail credit prediction data;
the feature primary screening module is used for carrying out primary screening on all categories comprising original retail credit prediction data and derived retail credit prediction data, namely all features, so as to obtain features after primary screening;
The primary screening data conversion module is used for judging a conversion mode of the features after primary screening so as to confirm that one of a WOE conversion mode, a dummy feature conversion mode and a continuous conversion mode is adopted for carrying out feature conversion, and carrying out feature conversion by adopting an optimal mode for judging each feature after primary screening;
The feature fine screening module is used for carrying out deep screening on the features subjected to the primary screening after feature conversion so as to obtain the features subjected to the fine screening;
A credit violation probability modeling module for model construction by selecting a logistic regression mode for the probability relation between the feature combination after fine screening and the credit violation, and confirming the mode for calculating the credit violation probability,
The data sample acquired by the data acquisition module is a customer sample using the consumed credit business.
55. The apparatus of item 54, wherein the apparatus performs the steps of the method of constructing a retail credit risk prediction model of any one of items 44 to 53.
56. A system for constructing a retail credit risk prediction model for a consumer credit service, the system comprising: a memory, a processor, and a program stored on the memory and executable on the processor to construct a retail credit risk prediction model method, the program to construct a retail credit risk prediction model method when executed by the processor implementing the steps of constructing a retail credit risk prediction model method as recited in any one of items 44 to 53.
Effects of the invention
The method and the system for constructing the retail credit risk prediction model combine a large number of samples of a large financial institution when constructing the model, deeply process and derive the original data obtained by the samples, and construct the retail credit general scoring model according to the self characteristics of the original data and the derived data and the strong financial attribute information which is difficult to acquire in the market and is contained in the data by utilizing an advanced statistical analysis method and an interpretable machine learning technology.
In addition, the method firstly utilizes a decision tree method to reasonably classify the samples before constructing the model, and constructs a sub-classification model based on the classified samples. The financial model is constructed by combining the decision tree, samples can be effectively classified according to executable categories, so that credit risk characteristics of different guest groups are better covered, and the problem that the model lacks subgroup representativeness due to the fact that all samples are used for constructing the model is avoided.
Further, when the risk prediction model is constructed, the model features are primarily screened by adopting a step-by-step discriminant analysis method, so that the overall efficiency of model development is fully improved. The progressive discrimination method can better discriminate more important characteristic variables under the same dimension, greatly reduces the workload of judging screening one by one according to the variable trend by a developer in the next step, fully improves the model development efficiency on the premise of not influencing the overall model effect, and ensures that the primary screening of the variables is more efficient and accurate.
The application combines three modes of continuous conversion, WOE conversion and dummy feature conversion in model construction, reprocesses partial features after initial screening, combines the advantages and disadvantages of the three conversion modes, designs a conversion judging method in an original way, and selects the optimal feature conversion mode according to parameters such as feature data attribute, deletion rate, concentration degree and the like assisted by judgment on business logic.
In addition, the method and the system for calculating credit risk probability or scoring credit risk constructed by the model of the application, namely a Beta model (Beta series scoring card) (comprising Beta 1-Beta 6 sub-models or sub-scoring cards), are characterized in that the model is firstly split based on a decision tree, so that the credit risk is calculated, firstly, a customer sample is effectively subjected to decision classification and is selected to be most suitable for the sub-model or the sub-scoring card for processing, and meanwhile, the model construction method used in the sub-model or the sub-scoring card is the method of the application, so that the technical defects of over fitting existing when a model is constructed by a complete WOE variable and the technical defects of being incapable of well adapting a split type variable existing when the model is constructed by a complete continuous variable are avoided in the construction process, and the method has an effect superior to the existing model in the aspect of predicting the credit risk.
The application is suitable for credit service, and is mainly aimed at the scene of the consumption credit service, and is based on analysis and design aiming at 1.3 billions of clients applying for registering the credit service, and particularly, model development is carried out by utilizing guest group samples applying for registering the consumption credit service, so that a credit service retail risk prediction model more suitable for financial institutions such as consumption financial companies, small credit companies and the like is constructed more pertinently. Aiming at the defects that the number of modeling samples is small, the data source is single, the data dimension homogeneity is high and the data with weak financial properties is mostly used in the existing scoring model, so that the deviation of a prediction result is large.
Drawings
FIG. 1 example 1 of the present application is an age distribution of an initial sample population for modeling.
Fig. 2 is a graph illustrating the result of the model discrimination effect of embodiment 1 of the present application.
FIG. 3 shows a typical flow chart for classifying all users of a sample based on a decision tree.
Fig. 4 shows an example of a binning confirmation of whether the data meets business trends.
Detailed Description
Credit risk is the risk that the borrower will have reduced or no way to perform on the borrowing contract due to economic capacity changes, rather than the risk of default due to intentional fraud by the borrower. Credit violations occur in various types of retail credit business scenarios, which are particularly relevant to borrower personal economies. The reasons for the borrower to overdue credit can be largely categorized into 4 categories: 1. the credit history is short, and the borrower has less experience in managing the financial condition; 2. the borrower forgets to repay temporarily; 3. excessive borrowers of this type have relatively low repayment capacity due to the large number of liabilities; 4. such borrowers are subject to significant negative factors that have a long-term impact on repayment capabilities due to significant factors such as reduced income, lost business, or divorced. The above reasons are more or less overdue for borrowers, and more serious default conditions may be caused, and the credit risk score is to mine the inherent mathematical relationship between various historical information of the clients and the probability of occurrence of future default, and convert the relationship into a score to quantify the probability of occurrence of default.
The credit risk score in the prior art mainly uses credit history data and multi-head data (multi-head data refers to behavior statistical data of borrowers for providing borrowing demands to a plurality of financial institutions, and generally considers that the larger the number of times of short-time multi-head is, the larger the future overdue probability is), so that payment behavior and payment willingness of the borrowers are reflected. The data of the sample used for constructing the model not only comprises the transaction behavior information of the credit service, but also increases the asset data which are difficult to acquire generally, and the application can comprehensively measure the repayment capacity, the personal qualification and the like of the borrower while reflecting the payment behavior and the payment willingness of the borrower, thereby providing more accurate prediction results.
< General description of model construction method >
In particular, the present application relates to a method of constructing a retail credit risk prediction model, comprising: a data acquisition step of acquiring raw retail credit prediction data of a sample used for constructing a model; a data deriving step of processing derived retail credit prediction data based on the original retail credit prediction data; a feature preliminary screening step of preliminarily screening all categories, i.e., all features, including the original retail credit prediction data and the derived retail credit prediction data to obtain preliminarily screened features; a preliminary screening data conversion step of judging a conversion mode of the preliminarily screened features to confirm that one of a WOE conversion mode, a dummy feature conversion mode and a continuous conversion mode is adopted to perform feature conversion, and performing feature conversion by adopting a judged optimal mode for each preliminarily screened feature; a feature fine screening step, namely performing deep screening on the features subjected to the primary screening after feature conversion to obtain the features subjected to the fine screening; and a credit violation probability modeling step, wherein a mode of logistic regression is selected for the combination of the characteristics after fine screening and the probability relation between the credit violations, so that a model is built, and a method for calculating the credit violation probability is confirmed.
In a specific embodiment, the method for constructing a retail credit risk prediction model according to the present application may further comprise a sample selection step, before the data acquisition step, for screening all users to obtain samples for model construction before the data acquisition step. Those skilled in the art understand that based on the total amount of users and the data situation that can build the model, those skilled in the art can choose whether to cull user data that does not conform to the built model. In addition, one skilled in the art may first select a part of users as samples, and then increase the modeled samples according to the corresponding rules.
< Sample selection step >
In a specific embodiment, the sample selection step is to screen all users to obtain samples for model construction prior to the data acquisition step. Specifically, the sample selection step includes classifying all users of the sample based on a decision tree, the classification basis including but not limited to: whether a user is a customer who has applied for registering credit services at a financial institution; whether a certain user belongs to a customer registered in a financial institution and not having a credit application; a geographic area to which a certain user handles business; whether a user has occurred a financial institution risk event (e.g., whether a default is currently occurring, whether a house is currently on hold); whether a user holds a credit card issued by a financial institution and/or whether the credit card is in continuous use and/or whether credit cards or personal loans are being used for circulation of credit.
Further, all the users of the sample are classified based on the decision tree, and the analysis sample is divided into whether the credit service is registered in the related bank or not because the analysis sample is required to have the performance variable during modeling, namely, the users can be divided into two parts of the applied registration and the non-applied registration.
Credit business is basic and important property business of banks, and benefits are obtained by paying bank loans to collect principal and interest and deducting cost. In general, banking credit businesses are an important means of banking profit, and can be classified into legal credit businesses and personal credit businesses in terms of classification of banking credit businesses. Wherein the corporate credit business includes project loans, mobile funds loans, small business loans, real estate business loans, etc.; the personal credit business includes a personal housing loan, a personal consumption loan, a personal business loan, and the like.
The Decision Tree (Decision Tree) is a Decision analysis method for evaluating the risk of an item and judging the feasibility of the item by constructing the Decision Tree to obtain the probability that the expected value of the net present value is greater than or equal to zero on the basis of the occurrence probability of various known situations, and is a graphical method for intuitively applying probability analysis. The application classifies samples for modeling based on the characteristics of large financial institution data, after dividing the samples into two parts of applied credit service samples and non-applied credit service, the samples of applied credit service can be further classified into two types of new clients and non-new clients according to whether the samples of applied credit service are new clients or not, wherein the decision of whether the samples of applied credit service are new clients or not can be determined based on the needs of technicians in the field, for example, users applying credit service within m months are decided to be new clients, and clients applying credit service for more than m months are non-new clients.
In a specific embodiment, m may be set to, for example, 3, may be set to, for example, 6, may be set to, for example, 9 or 12, and m may also be referred to as the account number in the present application.
It is well understood by those skilled in the art that other splitting methods may be selected when splitting the sample, for example, all samples may be split into a customer with a credit card and a customer group without a credit card at the time of splitting, and the customer sample splitting method at the time of model construction may be fully considered and performed according to the modeling requirement.
In the application, a consumer credit group sample is firstly selected for modeling, the characteristics of the sample can be effectively screened by adopting the sample group, the credit (including credit cards, consumer credits and other credits) and asset characteristics of consumer credit customers are mined, and a credit scoring method which is more suitable for financial institutions taking the consumer credit as a main business type is constructed. The model constructed by the sample can provide a prediction basis for a consumer finance company and a small credit company better.
< Original retail Credit data >
In constructing the model for the present application, about 120 kinds of original retail credit prediction data (i.e., 120 kinds of basic features) can be obtained based on basic data of a financial institution, and retail credit risk points are split or processed to the greatest extent possible on the premise of optimal effect according to different dimensions, so that 3000 derivative features are generated.
When the model is built, firstly, credit card application and behavior data, personal loan application and behavior data, personal financial property transaction data and personal customer information data of all retail customers in 4 years are initially collected during the model building of the application based on all historical data of a large financial institution, and 7.3 hundred million people are in total, wherein each person has corresponding data to be processed every month, so that a data system for building the model of the application is comprehensive and has very large data volume, a modeling method is considered during the model building based on the data system, otherwise, the special sample group needing to be concerned is covered in the huge data volume and cannot be effectively identified, and the computer program is slow to run or even cannot run, so that the most suitable prediction model cannot be accurately built.
In the data acquisition step, the raw retail credit prediction data of the sample for constructing the model obtained includes: credit card type base data, which is data that is fully available during sample-based credit card creation and use (i.e., credit card application and behavioral data of retail customers); the personal loan-type base data, which is all available data based on the loan application situation and the usage behavior of the sample (i.e., personal loan application and behavior data), the customer base information-type base data, which is data based on the properties of the sample itself but not directly related to the behavior at the financial institution (i.e., personal customer information data), or the personal financial property-type base data, which is all other financial property and financial transaction-type data (personal financial property transaction data) of the sample not related to credit cards and loans at the financial institution.
In one particular embodiment, the credit card type base data includes, but is not limited to: credit card billing, credit card cash withdrawal, credit card stage status, credit card generated interest and other information in different dimensions (such as time dimension, space dimension, frequency dimension). The above basic data is not limited to the specific types listed, and as credit card business changes, those skilled in the art can also fully cover the newly appeared data types in implementation, namely, all types of available data in the process of creating and using the credit card based on the user can be used as credit card basic data.
In one particular embodiment, the personal loan class base data includes, but is not limited to: information such as an account of the personal loan, overdue condition of the personal loan, balance of the personal loan, total amount of the personal loan, refund of the personal loan, actual refund, and the like. The above basic data is not limited to the specific types listed, and as the loan business of the individual varies, those skilled in the art can also fully cover the newly appeared data types, namely, the types of all available data based on the loan application situation and behavior of the user, which can be used as the basic data of the loan class of the individual.
In a specific embodiment, the basic data of the customer basic information class only comprises: the customer base information includes sex, age and administrative area information to which the business belongs. The above basic data is not limited to the specific categories listed, and with the development of customer situations and social relationships, those skilled in the art can also fully further cover other or new data types in implementation within the scope of the service application scenario, that is, all kinds of data based on the attributes of the user sample itself, but not directly related to the behaviors at the financial institution, can be used as the basic information base data of the customer.
In one particular embodiment, the personal financial asset class base data includes, but is not limited to: AUM (i.e., asset management size), deposit, financing, and shipping payroll information. The above basic data is not limited to the specific categories listed, and as the financial properties change, those skilled in the art can also fully and further cover the emerging data types in practice, i.e., all other financial properties and all types of financial transaction type data not related to credit cards and loans at the financial institution based on the sample, can be used as personal financial property type basic data.
In one particular embodiment, the raw retail credit forecast data for the sample taken to construct the model is based on more than 120 data types (i.e., base variables or base characteristics) taken by 7.3 million people, including but not limited to: information such as account, overdue, balance, amount, response, actual return and the like under different dimensions (such as time dimension, space dimension and frequency dimension) of credit card bill, credit card cash taking condition, credit card stage condition, interest generated by credit card and the like; information such as an account of the personal loan, overdue condition of the personal loan, balance of the personal loan, total amount of the personal loan, refund of the personal loan, actual refund, and the like; client basic information including sex, age and administrative region to which the business belongs; AUM (i.e., asset management size), deposit, financing, and shipping wages.
< Derived retail credit forecast data >
In the present application, in the data deriving step, the derived retail credit prediction data is processed based on the original retail credit prediction data, which is the data obtained by processing the collected original retail credit prediction data based on the time dimension, the space dimension, the frequency dimension, and the statistical information dimension.
In one embodiment, the derived retail credit prediction data includes, but is not limited to: derived retail credit prediction data processed based on sample relationship lengths, derived retail credit prediction data processed based on time interval class variables, derived retail credit prediction data processed based on sample behavior frequency levels, derived retail credit prediction data processed based on sample current time point conditions, derived retail credit prediction data processed based on sample sustained behaviors, or derived retail credit prediction data processed based on statistical information dimensions. For example, month data of the customer may be acquired, and processing may be performed based on the month data. Processing based on statistical information dimensions in the present application includes taking the maximum, minimum, average, etc. of the data to describe the data situation.
In a specific embodiment, for example, from a time dimension, for example, a customer relationship length class variable: such as customer account opening time, customer maximum account age, etc., as a category of derived retail credit prediction data, i.e., as a derived feature or derived variable.
In a specific embodiment, for example, starting from the time dimension, consider the case of a time interval, such as the number of months from the current time point of the customer's last payment, the number of months from the current time of the customer's last overdue, etc., as derivative features or derivative variables.
In a specific embodiment, for example, from the frequency level, consider the behavior frequency level variable: if the number of repayment > N in the last X months of the client, the number of the limit usage > N in the last X months of the client and the like are used as derivative features or derivative variables, X and N are not limited, and the service logic can be ensured to be a positive integer above any 0.
In a specific embodiment, consider the current time point variable, for example, from the time dimension: the current month limit of the customer, the current month balance of the customer and the like are used as derivative characteristics or derivative variables.
In one particular approach, from the time dimension, a continuous behavior variable is considered, such as the maximum number of consecutive times that the customer has exceeded > N for the last X months, the number of times that the customer has exceeded > N for the last X months, etc., as a derivative feature or derivative variable.
In a specific manner, from the dimension of the statistical information, a statistical class variable, such as the maximum expiration number of the last X months of the customer, the average credit usage within the last X months of the customer, etc., is considered as the derivative feature or derivative variable.
It will be apparent to those skilled in the art that the methods of processing the derivative variables described above are merely illustrative and may be arbitrarily chosen. For example, 120 types of base data may be processed into more than 3000 derived data. In the present application, there are sometimes cases where data, data types, variables or characteristics are mixed, and those skilled in the art can understand them based on the common knowledge of statistics.
In the application, the derivative data can be derived data obtained by simple processing of original retail credit prediction data or derived data obtained by complex processing, and the current variables such as current month wages and current month balances can be directly used after the data are summarized. Complex process derivative data is based on time slicing and logic processing based on current class variables, and derivative variables such as maximum payroll of approximately 3 months and minimum balance of approximately 12 months can be generated.
< Feature preliminary screening >
In the primary screening data conversion step, the conversion mode of the features subjected to primary screening is judged based on the concentration degree and the data type of the features subjected to primary screening. The primary screening data conversion step based on the judgment of the concentration and the data type comprises the following steps: classifying the data types of each feature to classify each feature into character type variable and numerical variable, performing primary screening data conversion on the character type variable by adopting a dummy feature conversion mode, and further classifying the numerical variable, wherein the process of classifying the numerical variable comprises the following sub-steps: if the value of the numerical variable is less than n, performing primary screening data conversion by adopting a WOE conversion mode, if the value of the numerical variable is more than n, further judging that if the conversion is that the continuous variable has more values and the concentration degree of a single value is more than m%, adopting the WOE conversion mode, and if the concentration degree of the single value is less than or equal to m%, adopting the continuous conversion mode, preferably, n and m are positive integers, wherein n=5-10, and m=90-99.
For example n is 5, 6,7, 8, 9 or 10 and m is 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99.
Specifically, in the application, a plurality of different feature preliminary screening modes are adopted to screen a plurality of features, so that the screening can be effectively carried out on the features with the largest dimension. Existing credit risk scores are often feature-prescreened using miss-rate, concentration, and information value IV.
In a specific embodiment, the features are screened based on the data loss of each feature of the sample used for constructing the model, and the deletion rate screening is generally performed by deleting a variable having a deletion rate of more than 90% or 91% or 92% or 93% or 94% or 95%, for example, features having a data deletion rate of more than 95% of the features may be deleted, or features having a data deletion rate of more than 90% may be deleted.
In a specific embodiment, features are screened based on the condition that a single value of a certain feature sample is too high, and in concentration screening, variables with the single value accounting for 99% or 98% or 97% or 96% or 95% or more are generally considered to be deleted, for example, features with the single value exceeding 99% can be removed, and features with the single value exceeding 95% can be removed.
In one embodiment, the information IV value for each feature is calculated to perform a preliminary screening of the feature. The IV value can be used for measuring the feature prediction capability in IV screening, and the larger the IV value is, the stronger the feature prediction capability is, and the calculation method of the single feature IV value is as follows:
wherein k is the number of packets after the feature is discretized; y i is the number of outstanding clients in the i-th group; y s is the total number of customers without violations; n i is the total number of default clients in the i-th group; n s is the total number of offending customers. The quantization index of IV has the following meaning: the characteristic is extremely weak when the calculated value of IV is less than 0.02, is weak when the calculated value of IV is less than 0.1 or more than 0.02, is good when the calculated value of IV is less than 0.3 or more than 0.1, and is strong when the calculated value of IV is more than 0.3.
Of course, the calculated value deletion threshold value of IV may be set to 0.03, 0.04, 0.05, or the like.
In the model construction method of the present application, the step of performing the feature preliminary screening using the deletion rate, the concentration degree, and the information value IV may be performed in any order, for example, the screening may be performed based on the deletion rate, the screening may be performed based on the concentration degree, and the feature preliminary screening may be performed based on the information value IV. Or screening based on concentration, screening based on the deletion rate, and finally performing feature primary screening based on the information value IV. Or screening based on the deletion rate, screening based on the information value IV, and finally performing feature primary screening based on the concentration degree. Or screening based on the information value IV, screening based on the deletion rate, and finally performing feature primary screening based on the concentration degree. Or screening based on the information value IV, then screening based on the concentration degree, and finally performing feature primary screening based on the deletion rate. Or screening based on concentration, screening based on the information value IV, and finally performing feature primary screening based on the deletion rate. The person skilled in the art can select based on the data condition of the sample, so that the characteristic with obvious defects in a certain aspect can be effectively removed based on the three methods, the data dimension can be effectively reduced, and the effect of model construction can be improved.
In a specific mode, the features after the screening by the utilization of the deletion rate, the concentration degree and the information value IV are respectively subjected to primary screening by adopting a gradual discrimination algorithm, and then the primary screening of the features is performed based on the coincidence condition of the risk characteristics of the features and the actual real results of the samples for model construction.
On the basis of the technical scheme, a gradual distinguishing method is introduced to improve the overall efficiency of model development, so that the primary screening of variables is more efficient and accurate. In actual data, there may be near distribution of good and bad samples on a certain variable, and the good and bad discrimination capability is weak. There may also be a class of variables, each able to distinguish good and bad samples well, but if all are included in the model, redundancy will be due to the duplication of the data dimensions covered by the variables. In contrast, the application uses a gradual discriminant analysis method, adopts WILKS' S LAMBDA value as the statistic criterion of admission and removal, and deletes the variables with weaker discriminant effect or redundancy in the data.
The gradual judging method adopts WILKS' S LAMBDA criterion to measure the strength of the characteristic capability and eliminates the characteristics which do not meet the set threshold value in the residual characteristics after three rounds of screening. In the gradual discriminating process, firstly, the variable with the strongest discriminating capability is added, along with the gradual increase of the variable in the model, the discriminating capability of the variable introduced earlier can also change, if the discriminating capability of a certain variable in the model is smaller than a threshold value, the variable is removed, and then the process is repeated until the variables contained in the model all meet WILKS' S LAMBDA similarity ratio criterion and other variables do not reach the standard of entering the model.
In the application, the step of screening by using a step-by-step discrimination method is a more critical step. The models involved in the present application use a huge amount of data and have numerous data dimensions for the purpose of capturing as much as possible all credit risk points, so variable screening must be performed to further reduce the time cost of model development. In the existing financial risk model evaluation, variable importance methods are generally adopted to perform variable deletion, such as calculating the importance of variables through algorithms such as a base index, an information entropy and the like, selecting variables with the front importance, and rarely using a modeling method for screening by a gradual discrimination method. Compared with a variable importance screening scheme commonly used in the industry, the method can retain a large number of variables with relatively weak importance and relatively independent information dimension.
In one particular approach, more than 3000 derived data may be processed based on, for example, 120 types of base data. Through three rounds of screening of the deletion rate, the concentration degree and the information value IV, the characteristics with about 20-30% of poorer data can be deleted, but more than 2000 characteristics can still be reserved, and if the follow-up variable fine screening is performed based on all the characteristics, the development efficiency is seriously affected. Therefore, the step-by-step discrimination is finally determined to be used as the optimal selection through comparison experiments of a plurality of different variable pruning schemes.
Aiming at the important features screened by the gradual discrimination algorithm, the further screening of the features is carried out based on the risk characteristics (system preset) of various risk points and the actual bad account rate condition of the sample. Judging whether the actual bad account rate distribution situation of the residual features accords with the business trend, and eliminating the features of the sample that the actual bad account rate distribution does not accord with the business trend. Specifically, (1) dividing the residual target features into 10-20 boxes according to the dividing points, and calculating the median of the values of each box and the corresponding bad account rate. (2) The change rate (slope) is calculated using the median of the values of each bin and the preceding bin and the corresponding bad account rate. (3) And counting the number of the boxes which are larger than 0 and the non-0 boxes (not larger than 0) in the change rate between two adjacent boxes, and calculating the percentage of the number of the boxes which are larger than 0 in the change rate to the number of the non-0 boxes in the change rate. (4) And acquiring the risk characteristic preset by the characteristic, and according to the percentage of the number of boxes larger than 0 in the calculated change rate to the number of boxes other than 0 in the change rate, judging whether the actual bad account rate change trend is approximately consistent with the business logic bad account rate change trend. The criteria for approximate agreement are as follows: first, if the business logic trend of this feature should be: as this feature value increases, the bad account rate increases (e.g., credit line usage, etc.). And in the module, eliminating the characteristics, of which the number of boxes which are larger than 0 in the calculated slope accounts for less than 70 percent of the non-0 boxes, namely eliminating the characteristics of which the characteristics show no business trend. Second, if the business logic trend of this feature should be: as this feature value increases, the bad account rate decreases (e.g., deposit amount, etc.). And in the module, eliminating the characteristics, of which the number of boxes which are larger than 0 in the calculated slope accounts for more than 30 percent of the non-0 boxes, namely eliminating the characteristics of which the characteristics show no business trend.
For example, the following table gives an example of binning, in which feature values are split into 10 bins, the range of values for the bins is also summarized in the following table, while a graph of the feature bin bad account rate as a function of the number of feature bin medians is plotted based on the data in the following table, as shown in fig. 4. In the example of fig. 4, the slope is greater than 7 for a total of 0 and less than 2 for a total of 0. By the method, the features can be further screened based on the risk characteristics (preset by the system) of various risk points and the actual bad account rate condition of the sample. The feature screening is further carried out by using the box division method, so that the features most conforming to the business development trend can be effectively screened out, and the features suitable for modeling are further obtained.
< Conversion of eigenvalues >
The original feature processing of the existing credit risk scoring method mainly adopts a mode of WOE conversion (multi-classification) and dumb feature conversion (bi-classification) to carry out discretization processing on continuous features (such as age, account age and the like). The WOE conversion mode is an optimal binning scheme based on modeling samples, continuous features are discretized according to optimal cutting points, and data of good and bad samples are embedded into WOE values, so that the performance of the WOE conversion mode is better in model construction, but the binning result can be excessively fitted with the modeling samples, so that the problem that the model effect is seriously reduced when the WOE conversion mode is applied to the whole (the generalization capability is poor) is solved. Meanwhile, original features falling into different boxes are converted into single values corresponding to the boxes by normalization operation used in the boxes, so that the risk distinguishing capability of people falling into the same interval is lost. The conversion mode of the dummy feature is mainly applied to the grouping feature, and has the advantages that the quality difference between different values of the feature can be eliminated, but the continuous variable is abnormally complex and redundant in processing.
In the model construction method, the conversion mode of the features after preliminary screening is judged to confirm that one of the WOE conversion mode, the dummy feature conversion mode and the continuous conversion mode is adopted for carrying out feature conversion, and the optimal mode of judgment is adopted for carrying out feature conversion on each feature after preliminary screening.
The primary screening data conversion step based on the judgment of the concentration and the data type comprises the following steps: classifying the data types of each feature to classify each feature into character type variable and numerical variable, performing primary screening data conversion on the character type variable by adopting a dummy feature conversion mode, and further classifying the numerical variable, wherein the process of classifying the numerical variable comprises the following sub-steps: if the value of the numerical variable is less than n, performing primary screening data conversion by adopting a WOE conversion mode, if the value of the numerical variable is more than n, further judging that if the conversion is that the continuous variable has more values and the concentration degree of a single value is more than m%, adopting the WOE conversion mode, and if the concentration degree of the single value is less than or equal to m%, adopting the continuous conversion mode, preferably, n and m are positive integers, wherein n=5-10, and m=90-99.
Specifically, taking a character type variable as an academic example, the value of the characteristic variable can be primary school, middle school, university, research student, etc. For a numerical variable, if the numerical variable is the past 3 months overdue month number, the value is 0, 1, 2, 3.
In the present application, n may be 5, 6, 7, 8, 9 or 10, and m may be 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99.
In a specific embodiment, m is selected to be 5 and n is selected to be 95.
Specifically, the manner of WOE conversion is: searching an optimal cutting point of the feature, dividing the value domain of the original feature to obtain a plurality of sub-boxes, calculating WOE conversion values corresponding to each divided box according to the quality condition in each sub-box, outputting the WOE conversion values, dividing the original feature according to the sub-box result, and outputting the WOE conversion values. For each bin, the WOE value is calculated as follows:
Wherein y i is the number of outstanding clients in the i-th group; y s is the total number of customers without violations; n i is the total number of default clients in the i-th group; n s is the total number of offending customers.
The WOE binning process is described by way of example with respect to age. The original characteristic age comprises a numerical value of 18 to 50 years old, 5 sub-boxes are obtained after sub-boxes, namely 18 to 24 years old, 25 to 30 years old, 31 to 35 years old, 36 to 42 years old and 43 to 50 years old respectively, then the WOE conversion value of each sub-box is calculated according to the number of good or bad clients of each box, and finally each piece of user data divided into each sub-box is output according to the corresponding conversion value of each sub-box, for example, the corresponding WOE conversion value of 18 to 24 sub-boxes is output at 23 years old, and the corresponding WOE conversion value of 43 to 50 years old sub-boxes is output at 46 years old.
The conversion mode of the dummy features is as follows: and converting the single classification feature into dummy features with the same quantity according to the quantity of the included numerical values, wherein if a certain client belongs to the generated dummy feature corresponding value, the corresponding dummy feature value is 1, and the rest dummy feature values are 0.
The manner in which dummy features are converted is described by way of example in terms of gender. The original features include: after the dummy features are converted, two dummy features of 'sex-male' and 'sex-female' are generated, if the sex of the client is male, the sex-male 'is marked as 1, and the sex-female' is marked as 0. Taking the learning example, the original features include: big proprietary and below, big book, master and above, generating three dummy features of 'academy-big proprietary and below', 'academy-big book', 'academy-master and above' after dummy feature conversion, if the client's academy is the great book, it is marked as 0 in' academy-great expert and below, 'academy-great book' is marked as 1, and 'academy-great person and above' is marked as 0.
The continuous conversion mode is as follows: the original features are subjected to continuous conversion in a plurality of ways (the ways of continuous conversion include but are not limited to, directly selecting original values, calculating squares of the original data, calculating square roots of the original data, calculating cube roots of the original data, and calculating natural logarithms of the original data). And calculating the correlation coefficient r (Correlation Coefficient) between the continuous converted characteristic value and the overdue label, selecting the conversion mode with the maximum absolute value of the correlation coefficient for conversion, and outputting the conversion mode corresponding to the original characteristic. The calculation formula of the correlation coefficient is as follows:
Wherein Σ is the sum symbol in mathematics; n is the total number of observations; x i is the converted value of the original feature of the ith observation after continuous conversion; an average value of the conversion values for this purpose; wherein y i is the ith observation that represents a classification feature of whether or not to violate; /(I) For this purpose, the average value of the classification characteristic. The closer the absolute value of the correlation coefficient is to 1, the more relevant the converted conversion value is to the default condition, and the better the effect of the conversion mode is. The quantization index meaning of the correlation coefficient r is as follows: when the absolute value of the correlation coefficient is 0 or more and less than 0.3, a low degree of correlation is represented, when the absolute value of the correlation coefficient is 0.3 or more and less than 0.8, a medium degree of correlation is represented, and when the absolute value of the correlation coefficient is 0.8 or more and less than 1, a high degree of correlation is represented.
In the model construction of the application, aiming at confirming the feature adopting the continuous conversion mode, an optimal conversion method is selected to carry out continuous feature conversion of the feature based on the correlation between the feature and credit violations under different continuous conversion modes, and preferably, the continuous feature conversion is carried out by adopting the modes of directly selecting an original value, calculating the square of original data, calculating the square root of the original data, calculating the cube root of the original data or calculating the natural logarithm of the original data.
The continuous conversion mode is described by way of example with respect to age. The original characteristic age comprises a numerical value of 18 to 50 years old, the original value, square root, cube root, natural logarithm and the like of the age are obtained through continuous conversion, then the absolute value of the WOE value and the correlation coefficient of each conversion value and a quality label (prediction result) is calculated, and a conversion mode with the maximum correlation coefficient absolute value is selected for conversion and then output. If the cube root of the age characteristic has the largest absolute value of the correlation coefficient with the quality label compared with other conversion modes, outputting the cube root of the age as a conversion value.
In the prior art, one feature conversion mode of WOE and dummy features is mainly selected for model construction. For example, in the chinese patent CN112686749B, the characteristic value is converted by WOE method. The WOE conversion mode is an optimal binning scheme based on modeling samples, and continuous features are discretized according to an optimal cutting point, so that the performance of the WOE conversion mode is better when a model is built, but the binning result can be excessively fitted with the modeling samples, so that the model effect is seriously reduced (the generalization capability is poor) when the WOE conversion mode is applied to the whole. Meanwhile, original features falling into different boxes are converted into single values corresponding to the boxes by normalization operation used in the boxes, so that the risk distinguishing capability of people falling into the same interval is lost.
The conversion mode of the dummy feature is mainly applied to the grouping feature, and has the advantage of eliminating the difference between different values of the feature. Such as: for the industries where clients are located, the method is more suitable for the processing mode of the dummy features because the retail industry and the wholesale industry have no obvious advantages and disadvantages; there is a hierarchical difference between specialty and family, so although dummy features can be used, in practice the treatment with WOE is more appropriate.
On the other hand, the continuous conversion mode can avoid the situation of overfitting the modeling sample in the WOE conversion mode, has stronger generalization capability on the overall sample, and has fewer situations of entering a single value for most guest communities because interval mapping is not performed. But cannot be applied to features that are partially monotonically poor or discrete (e.g., occupation, job, etc.).
As described above, the technical solution of the present application is to develop a new way to combine three ways of continuous conversion, WOE conversion and dummy feature conversion, reprocess part of the features after careful screening, combine the advantages and disadvantages of the three conversion ways, and design a conversion judgment method originally, and select the optimal feature conversion way by assisting with judgment on business logic according to parameters such as feature data attribute, deletion rate, concentration degree, etc.
< Logistic regression and feature depth screening based on logistic regression >
In the existing credit risk score, for the requirement of model interpretability, a logistic regression model is mainly used for model development, and software can be used as follows: SAS, R, python, etc.
In a specific embodiment, the present application is based on SAS software for model development.
Specifically, the probability predicted to be default is fitted in logistic regression using a Sigmoid function:
Wherein Z is the linear combination of the model coefficient and the feature conversion value, and Z is defined as follows:
Z=α+β1x1+β2x2+...+βk-1xk-1+βkxk
The probability of predicted default is:
P=P(Y=1|x1,x2,x3,...,xk-1,xk)
the probability of the predicted default after fitting is:
From the above equation it can be further deduced that:
Substituting the Z value into the above formula can calculate the probability P predicted to be default.
The core of the logistic regression model construction is feature screening, wherein the feature screening steps are as follows: first, the features are screened in batches according to the deletion rate, the concentration degree and the information value IV. And secondly, screening all the remaining features one by one according to whether the features accord with the business trend, and reserving the features with correct business trend. For example: if the bad account rate of the guest group is found to be reduced along with the increase of the balance of the feature loan, the feature is considered to be not in accordance with the business trend. In credit business understanding, the higher the loan balance, the higher the customer's level of exposure to default risk (EAD) and the greater the risk. This feature is removed from the feature list at this point. Third, the stepwise regression function using logistic regression eliminates features that are not very important and highly correlated with other features. Fourth, the coefficient is screened according to the positive and negative signs of the training coefficient and the business trend of the feature conversion value, and the feature that the sign of the feature coefficient accords with the business logic is reserved. Defining 0 as a good customer and 1 as a bad customer in the Y tag, wherein for a feature (such as loan balance) with monotonically increasing bad account rate along with the increase of feature value, the training coefficient sign of the feature (such as loan balance) is positive; on the contrary, for a feature (e.g. deposit balance) whose bad account rate decreases monotonically with the increase of the feature value, the training coefficient sign should be negative, and if the training coefficient sign does not meet the above criteria, the training coefficient sign should be removed. Fifth, features with higher correlation are further eliminated by using a variance expansion factor (VIF), a correlation coefficient, and the like: for the variance expansion factors, the features with the highest VIF and greater than 4 are eliminated one by one; for the features with higher correlation, the features with lower IV values in the feature group with the highest correlation coefficient and larger than 0.80 are eliminated one by one. Sixth, the characteristic that the distribution difference of the group stability coefficient (PSI) is large at different time points so as to cause instability is removed, the characteristic of PSI >0.25 is directly removed, and the characteristic of PSI >0.1 is removed according to the influence of removing the characteristic on the model distinguishing capability.
The application screens the characteristics strictly following the rules, and ensures the interpretation and stability of the model and the distinguishing capability of good and bad clients.
The characteristic fine screening step of the application comprises the following steps: a first fine screening step of screening features based on a stepwise regression algorithm, a second fine screening step of calculating a variance expansion factor based on each feature and eliminating features with higher variance expansion factors to screen the features, a third fine screening step of analyzing whether the feature coefficients conform to the trend of the prediction result for credit violations based on a logistic regression on the features after the first fine screening step and the second fine screening step to further screen the features.
And a feature fine screening step 1, which is based on a stepwise regression algorithm and on an F test and T test method, wherein features are sequentially introduced from high to low on the basis of significance, and each time a feature is introduced, the selected features are tested one by one. The features originally introduced are rejected when they become no longer significant due to the introduction of the following features. This process is cycled until no features above the significance threshold are selected into the equation, or no features below the significance threshold are removed from the regression equation.
And (2) a feature fine screening step, which is based on a method for eliminating features with higher variance expansion factors, and further reduces multiple collinearity in the model.
And 3, comparing the risk characteristics (preset by the system) of various risk points with the positive and negative signs of the model training coefficients, judging whether the characteristic coefficients of the residual characteristics in the model in the step 3 accord with the business trend, eliminating the characteristics of the model coefficients not accord with the business trend, and iterating again. The specific embodiment of the feature fine screening module 3 is as follows: 1. for the characteristics of WOE class in the characteristic conversion mode, the corresponding model training coefficient should be negative, and the WOE conversion class characteristics of which the training coefficient is positive should be removed. 2. For the continuous conversion mode, if the feature value is increased in service logic, the bad account rate should be increased (such as credit line usage rate, etc.), the corresponding model training coefficient should be positive, and the continuous conversion type feature with the training coefficient being negative should be removed; if the feature value increases in service logic and the bad account rate should decrease (e.g. deposit amount, etc.), the corresponding model training coefficient should be negative, and the continuous conversion type feature with the training coefficient being positive should be removed. 3. For the dummy feature conversion mode, the bad account rate with the value of 1 and the value of 0 is required to be judged, if the bad account rate with the value of 1 is larger than the bad account rate with the value of 0, the coefficient is positive, otherwise, the coefficient is negative.
And a feature fine screening step 4, namely a data stability monitoring step, wherein the data stability monitoring step is used for evaluating whether obvious deviation occurs to the distribution situation of single features and integral scores of different points, the features of PSI >0.25 are directly removed, and the features of 0.25 PSI >0.1 are removed with cautions according to the influence of removing the features on the model distinguishing capability.
And (3) carrying out multiple iterations on the feature list based on the steps (3) and (4), stopping the iteration until no new feature is added or removed from the model finally, and obtaining a final feature list and a conversion value thereof. Through the steps, the modeling variable used for constructing the model of the application can be finally obtained.
The credit violation probability modeling step of the application substitutes the characteristics screened by the characteristic fine screening step into a Sigmoid function to carry out logistic regression calculation on the credit violation probability model.
In the prior art, the core of all scoring models constructed is whether the data used is representative. For the scoring models existing in the prior art, the sample size and bad labels of most models are smaller due to information safety and cost, the stability of the models cannot be guaranteed, and independent modeling of a certain type of clients cannot be achieved. Meanwhile, because the scoring model in the market is mostly multi-head lending data and weak financial attribute data (such as non-credit transaction data including intelligent terminal equipment data, social platform data, online shopping mall data and the like) in the information dimension in the modeling process, the asset condition and repayment capability of a customer cannot be accurately reflected. The data source used by the application is the full traffic data of a large bank, and the data magnitude of the modeling sample and the bad label is huge. According to the application, modeling samples are designed based on different sample groups, so that risk differences among clients can be distinguished more finely, and meanwhile, stability of data and a model is ensured based on cross-time verification and PSI verification. The application uses the personal credit history data and the asset data to carry out model development, thereby better embodying the repayment willingness and repayment capability of borrowers.
For the internal self-built scoring model of a part of small and medium-sized financial institutions, the accumulated historical data is insufficient to develop a credit risk model with stable data and strong distinguishing capability due to the fact that retail credit business volume is small or business development is late, so that the dependency on manual approval is high when the model is used for credit approval business. The efficiency limit of manual approval restricts the development of retail credit business, and meanwhile, the operation risk in the credit approval process is increased due to subjectivity of the manual approval. The model constructed by the application can assist the financial institutions to make digital decisions, enhance the approval accuracy and the approval speed, and reduce the adverse effects.
In the aspect of feature conversion, compared with the traditional WOE conversion mode, continuous features need to be subjected to rough box division and discretization according to data representation and experience of modeling staff, and the result of rough box division is greatly influenced by subjective factors of the modeling staff; and the process of discretizing the continuous features may cause too many clients to fall into the same interval, thereby resulting in a large number of calculated score single values. The application combines the continuous conversion mode, the WOE conversion mode and the dummy feature conversion mode to encode the original features, thereby reducing the influence caused by human factors and single values and enhancing the grading distinguishing capability.
< Method of calculating retail Credit Risk >
The application further relates to a method for calculating the retail credit risk of the sample to be tested by using the retail credit risk model constructed by the application.
In the present application, retail credit risk refers to the probability of credit violations for the sample under test. In one particular approach, the probability of credit violation refers to the probability of occurrence of a credit overdue, and specifically, for example, may be the probability of whether a credit overdue occurs for more than 60 days.
The application relates to a method for calculating retail credit risk, comprising: a data acquisition step of acquiring retail credit prediction data of a sample to be predicted; classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating credit violation probability; and a credit violation probability calculating step, wherein retail credit prediction data are substituted into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted.
In the present application, the sample to be predicted may be a sample used to build a model, i.e., the sample is already a customer of a financial institution holding credit services (including but not limited to credit cards or personal loan services) at the time of the model build, and the method of calculating retail credit risk using the present application may be used to calculate the potential retail credit risk in the future. In the present application, the sample to be predicted may be a sample not used for modeling, i.e. not already a customer of a financial institution holding a credit service at the time of modeling, but now already a customer of a financial institution holding a credit service, and the method for calculating retail credit risk according to the present application may be used for calculating the potential retail credit risk in the future. In the present application, the sample may also be a customer who has not, and now is, a model built in the past itself, a credit business held by the financial institution, but a customer who has already held other financial asset-like businesses, and the method of calculating retail credit risk of the present application may be used to calculate the potential retail credit risk in the future.
The application relates to a method for calculating retail credit risk, comprising: a data acquisition step of acquiring retail credit prediction data of a sample to be predicted; classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating credit violation probability; a credit violation probability calculating step of substituting retail credit prediction data into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted; and after calculating the credit violation probability, calculating the credit score of the sample to be predicted.
The step of calculating the credit score of the sample to be predicted is used for calibrating the calculated credit violation probability to a normalized score of 0-1000 points.
In the method of the present application, the retail credit prediction data includes raw retail credit prediction data for the sample to be predicted and derived retail credit prediction data is processed based on the raw retail credit prediction data. Wherein the description of the original retail credit prediction data and the derived retail credit prediction data are identical to the description of the model building portion.
In the present application, the step of classifying the sample to be predicted includes the following sub-steps: whether the sample to be tested is a customer who has applied for a credit registering service at a financial institution; whether the sample to be tested is a customer who has a credit overdue; whether the current credit utilization rate of the sample to be tested is more than or equal to a percent; whether the current asset size of the sample to be tested is larger than or equal to b. Classifying the samples to be predicted based on the above sub-steps to determine a sub-model for calculating the credit violation probability, and the order in which the above sub-steps are performed may be arbitrarily set.
In the present application, fig. 3 shows a flow chart for classifying samples to be predicted, and based on the classification of this step, a submodel for calculating the credit violation probability can be selected. The samples to be predicted are classified as shown in fig. 3 in the following order: firstly, judging whether the sample to be tested is a customer who has applied for registering credit business at a financial institution; then judging whether the sample to be tested is a customer with overdue credit; then judging whether the current credit utilization rate of the sample to be tested is more than or equal to a% (for example, a can be 50, 60,70,80,90,95,96,97,98,99,100 and the like); and then judging whether the current asset size of the sample to be tested is larger than or equal to b (for example, b can be 100 yuan, 200 yuan, 500 yuan, 1000 yuan, 5000 yuan, 10000 yuan, 10 ten thousand yuan, 100 ten thousand yuan, etc.). Using such step classification, the best fit sub-model for predicting the sample to be tested can be identified.
According to the application, after feature conversion is carried out on the retail credit prediction data, the feature conversion is substituted into the credit violation probability submodel to calculate the credit violation probability of the sample to be predicted, and the feature conversion step comprises the step of selecting a WOE mode or a continuous mode to carry out feature conversion based on the feature type of the retail credit prediction data which needs to be substituted into the credit violation probability submodel.
Methods for feature transformation for WOE mode or continuous mode are well known to those skilled in the art, and reference is made to the description of the application in the model building section for a specific transformation mode. However, the existing model in the field only adopts the WOE mode to perform parameter conversion.
The WOE conversion mode is an optimal binning scheme based on modeling samples, and continuous features are discretized according to an optimal cutting point, so that the performance of the WOE conversion mode is better when a model is built, but the binning result can be excessively fitted with the modeling samples, so that the model effect is seriously reduced (the generalization capability is poor) when the WOE conversion mode is applied to the whole. Meanwhile, original features falling into different boxes are converted into single values corresponding to the boxes by normalization operation used in the boxes, so that the risk distinguishing capability of people falling into the same interval is lost. The conversion mode of the dummy feature is mainly applied to the grouping feature, and has the advantage of eliminating the difference between different values of the feature. Such as: for the industries where clients are located, the method is more suitable for the processing mode of the dummy features because the retail industry and the wholesale industry have no obvious advantages and disadvantages; there is a hierarchical difference between specialty and family, so although dummy features can be used, in practice the treatment with WOE is more appropriate. The continuous conversion mode can avoid the situation of overfitting the modeling sample in the WOE conversion mode, has strong generalization capability for the overall sample, and has fewer situations of entering a single value for most guest groups because interval mapping is not performed. But cannot be applied to features that are partially monotonically poor or discrete (e.g., occupation, job, etc.). The application processes the initial feature by combining continuous conversion and WOE conversion, combines the advantages and disadvantages of the two conversion modes, designs a conversion judging module, and selects the optimal feature conversion mode by assisting with judgment on business logic according to parameters such as feature data attribute, deletion rate, concentration degree and the like. The technical scheme of the application develops a new way, combines three modes of continuous conversion, WOE conversion and dummy feature conversion, reprocesses part of the features which are carefully screened, combines the advantages and disadvantages of the three conversion modes, designs a conversion judging method, and selects the optimal feature conversion mode by assisting with judgment on business logic according to parameters such as feature data attribute, loss rate, concentration degree and the like. The feature conversion mode determined by the mode is substituted into the model for calculating the retail credit risk, so that the retail credit risk of the sample can be predicted more accurately.
The sub-model for calculating the credit violation probability is a model constructed based on the sample retail credit prediction data and the credit violation probability by adopting logistic regression based on the existing user population, namely, a model constructed by utilizing the method described by the application.
< Apparatus, system and computer storage Medium for calculating retail Credit Risk >
The application relates to a device for calculating retail credit risk, comprising: the data acquisition module is used for acquiring retail credit prediction data of a sample to be predicted; a module for classifying the sample to be predicted, for classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating a credit violation probability; a credit breach probability calculation module for substituting retail credit prediction data into a credit breach probability submodel to calculate a credit breach probability for the sample to be predicted.
The inventive means for calculating retail credit risk may perform the steps of the inventive method for calculating retail credit risk.
The application also relates to a system for calculating retail credit risk, which is characterized in that the system for calculating retail credit risk comprises: the system comprises a memory, a processor and the program for calculating retail credit risk stored in the memory and capable of running on the processor, wherein the program for calculating retail credit risk is executed by the processor to realize the steps of the method for calculating retail credit risk according to the application.
The application relates to a computer storage medium, characterized in that the computer storage medium has stored thereon a program for a method of calculating retail credit risk, which program, when being executed by a processor, implements the steps of the method of calculating retail credit risk according to the application.
All of the above description of the method for calculating retail credit risk may be fully applicable to the apparatus, system, and computer storage medium for calculating retail credit risk.
The method for calculating the retail credit risk can avoid the technical defects that the model is built by the complete WOE variable, the model is built by the complete continuous variable, and the model cannot be well adapted to the classified variable, so that the model can better cover the retail credit risk prediction requirements of various different types of passenger groups when the model is used for calculating the retail credit risk, and can meet the requirement of obtaining more accurate prediction results when the model is used for general scoring prediction of the credit risk.
Examples
Example 1 collection of modeling samples
In the process of construction of this embodiment, credit card application and behavior data, personal loan application and behavior data, personal financial property transaction data, and personal customer information data of all retail customers in the construction bank between 2017 and 2021 are collected for a total of 7.3 hundred million people. Modeling sample confirmation was performed by a professional model design scheme, 6.3 million data in 2019 was selected as an analysis sample, and 6.3 million clients were age-distributed as shown in fig. 1. Since the modeling requires the presence of performance variables, the analysis sample is divided into two parts, namely an applied registration credit service and an applied no registration credit service, and the sample size is 1.3 hundred million and 5 hundred million people respectively.
The specific Beta model scores the scenes of the consumption credit business, the analysis and design are carried out on 1.3 hundred million clients who have applied for registering the credit business when the model is designed, and the online result is applied to 7.3 hundred million people with or without application or other potential whole client groups after the model is developed.
The model design comprises: 1) Rule exclusion: for example, the customer data of the established, sold, no expression and special situation are excluded, and the modeling crowd after the exclusion is 248 ten thousand customers (the partial crowd is the customer group applying for registering the consumption credit business); 2) Time window setting: determining that the sample uses 2019 data as a modeling sample and uses a time range of 15 months in the future as a performance period of Y; 3) Sampling: sampling modeling was performed with a sample number of good/bad 5:1. And designing different subdivision schemes in a decision tree mode, confirming the final scheme of model subdivision by adopting a comparison method of father and son models, and further modeling each subdivision model respectively. The model subdivision scheme is designed by dividing modes of whether a client is applied for, whether the client is overdue, whether the credit utilization rate is greater than 100%, whether the current asset is greater than 500 yuan and the like, fully embodies control variables closely related to risk characteristics in the service development process, and can be comprehensively matched with various client groups on the market.
In this embodiment, the modeling sample is first divided into the following sub-model modeling samples based on the decision tree for the construction of the subsequent sub-models. When determining a modeling sample, the first layer of the decision tree is divided according to whether credit overdue exists currently or not, and is used for distinguishing historical repayment behaviors of user groups. In this embodiment, the classification of the users with the current overdue number not less than 1 (i.e. defined as the current overdue) is classified into a Beta5 model, and the users with the current overdue number=0 (i.e. defined as the current overdue) are classified into four models of Beta1, beta2, beta3 and Beta4 according to the current month credit usage rate and the current month asset level. (wherein "current month credit utilization rate" means the current month credit balance to line ratio; the "credit utilization rate < 100%" means the current month credit balance to line ratio <100%; "credit utilization rate > 100%;" means the current month credit balance to line ratio > = 100%). Specifically, for the user population with current overdue number=0 (i.e. current unducted), the user population is divided according to its current credit utilization rate into two cases, the current credit utilization rate <100% and the credit utilization rate > 100%; for the two cases, the division is further performed according to the current asset size, wherein the user crowd division with the current credit utilization rate less than 100% and the current asset size less than 500 yuan is classified as Beta1; dividing the user population with the current credit utilization rate less than 100% and the current asset size more than or equal to 500 yuan into Beta2; dividing the user population with the current credit utilization rate more than or equal to 100% and the current asset size less than 500 yuan into Beta3; the user crowd division with the current credit utilization rate more than or equal to 100 percent and the current asset size more than or equal to 500 yuan is classified as Beta4.
In addition, in this embodiment, in order to construct a model for a sample group without application for registration credit service, a user group with an account age less than 3 months is selected as an approximate guest group, and then the model training is performed by entering Beta6 subdivision.
Specifically, the sample size of the sub-guest group used for constructing the Beta1 subdivision model is about 32 Mo Zuo, the guest group is mainly a customer who has applied for credit business and is not overdue at present, the credit utilization rate is less than 100 percent, the current asset size is less than 500 yuan, and the model is constructed for predicting the probability of occurrence of overdue credit for 60 days and more of the crowd.
Specifically, the sample size of the sub-guest group used for constructing the Beta2 subdivision model is about 180 ten thousand, the guest group is mainly a customer who has applied for credit business and is not overdue at present, the credit utilization rate is less than 100 percent, the current asset size is more than or equal to 500 yuan, and the model is constructed for predicting the probability of occurrence of overdue credit for the crowd of the customer for 60 days or more.
Specifically, the sample size of the sub-guest group used for constructing the Beta3 subdivision model is about 6 ten thousand, the guest group is mainly a customer who has applied for credit business and is not overdue at present, the credit utilization rate is more than or equal to 100%, the current asset size is less than 500 yuan, and the model is constructed for predicting the probability of occurrence of overdue credit for 60 days or more for the crowd.
Specifically, the sample size of the sub-guest group used for constructing the Beta4 subdivision model is about 11 ten thousand, the guest group is mainly a customer who has applied for credit business and is not overdue at present, the credit utilization rate is more than or equal to 100 percent, the current asset size is more than or equal to 500 yuan, and the model is constructed for predicting the probability of occurrence of overdue credit for 60 days or more for the crowd.
Specifically, the sample size of the sub-guest group used for constructing the Beta5 subdivision model is about 15 ten thousand, the guest group is mainly the current overdue client who has applied for credit business, and the model is constructed for predicting the probability that the crowd thereof has overdue credit for 60 days or more.
Specifically, the sample size of the sub-guest group used for constructing the Beta6 subdivision model is about 3 ten thousand, the guest group is mainly a customer with the account age smaller than 3 months of the applied credit service, and the model is constructed for predicting the probability of the future occurrence of overdue credit of the sample group without the applied registered credit service for 60 days or more.
Based on the confirmed different guest group samples of each sub-model, historical data information of borrowers is obtained: 1) Credit cards, the basic field contains account, overdue, balance, amount, refund, actual refund and other information under different dimensions such as bill, cash withdrawal, installment, interest and the like; 2) The personal loan class comprises account, overdue, balance, amount, response, actual response and other information; 3) Customer base information including sex, age, and administrative area information to which the service belongs; 4) Personal financial assets, including AUM, deposit, financing and shipping wage information.
And splitting macroscopic credit risks according to various risk points in credit business and information dimension, time slicing and other modes. The variable derivation is carried out by a professional characteristic variable construction method, and the basic modes are as follows: 1) Customer relationship length class variable: such as the customer account opening time, the maximum account age of the customer, etc.; 2) Time interval class variable: such as the month number of the last repayment of the customer from the observation point, the month number of the last overdue of the customer from the present month, etc.; 3) Behavior frequency degree variable: such as the number of repayment > N in the last X months of the customer, the number of the limit use rate > N in the last X months of the customer, etc.; 4) Current time point variable: current month limit of the customer, current month balance of the customer, etc.; 5) Statistical value class variable: if the maximum overdue number of the client is X months recently, the average limit use rate of the client is X months recently; 6) Continuous behavior variable: such as the maximum continuous times of continuous overdue > N of the last X months of the customer, the continuous times of continuous > N of the repayment rate of the last X months of the customer, and the like. Finally, 3067 features with potential prediction capability for the overdue condition of the client, such as 'current payment amount', 'average overdue number of the last 6 months', 'credit card overdue number of the last 12 months', and the like, are derived. In the present application, the observation point refers to the time point when the sample is collected by the time of modeling, and currently refers to the sampling cutoff time, and the observation point and the current represent the same meaning.
Table 1 summary of the basic and derivative variables used in the examples
Example 2 feature preliminary screening
The 3067 features (variables) collected in example 1 were initially screened for potential predictability of customer overdue.
The preliminary screening for the 5 sub-model features of Beta1 through Beta5 divided in example 1 was performed as follows:
the first round of preliminary screening, first culled out the features with deletion rate exceeding 95% in the data collected in example 1, deleted 689 variables altogether, and the remaining 2378 variables.
And (3) the second round of preliminary screening, namely, eliminating the features with single values exceeding 99% in the remaining features aiming at 2378 features subjected to the first round of preliminary screening, eliminating 208 variables altogether, and remaining 2170 variables.
And thirdly, primarily screening, namely sorting 2170 features subjected to the primary screening of the second round according to feature values (the sorting method is specifically that each value is independently one box if the features are character type variables, sorting is carried out according to the size of the values from small to large if the features are numerical type variables), equally dividing 10-20 boxes according to the dividing points, calculating a feature IV value, and rejecting the feature IV value if the feature IV value is lower than 0.02, wherein in the embodiment, because the variable attribute is strong financial and asset type variables, the prediction effect is strong, rejecting the variables with the IV value lower than 0.05, totally rejecting 706 features, and remaining 1464 feature variables.
And a fourth round of preliminary screening, wherein the 1464 features subjected to the third round of screening are subjected to further screening based on a step-by-step discrimination algorithm, and 208 important feature variables can be rapidly screened out through the third round of screening. The step-by-step discrimination method is described below, and there may be little difference in the average value of different categories on a certain variable in actual data, in which case the effect of classifying using the variable will not be good; there are also a class of variables that do distinguish well between different classes in the data when considered independently, but that may appear redundant when included into the model variables. Therefore, the embodiment creatively adopts a gradual discrimination method, better screens out more important characteristic variables under the same dimension, greatly reduces the workload of judging screening one by one according to the variable trend in the fifth round of screening, and fully improves the model development efficiency on the premise of not affecting the overall model effect.
The gradual judging method adopts WILKS' S LAMBDA criterion to measure the strength of the characteristic capability and eliminates the characteristics which do not meet the set threshold value in the residual characteristics after three rounds of screening. In the gradual discriminating process, firstly, the variable with the strongest discriminating capability is added, along with the gradual increase of the variable in the model, the discriminating capability of the variable introduced earlier can also change, if the discriminating capability of a certain variable in the model is smaller than a threshold value, the variable is removed, and then the process is repeated until the variables contained in the model all meet WILKS' S LAMBDA similarity ratio criterion and other variables do not reach the standard of entering the model.
And a fifth round of preliminary screening, wherein the 208 important features subjected to the fourth round of screening are subjected to further screening based on risk characteristics (system preset) of various risk points and actual bad account rate conditions of samples.
Judging whether the actual bad account rate distribution situation of the residual features accords with the business trend, and eliminating the features of the sample that the actual bad account rate distribution does not accord with the business trend. In particular the number of the elements,
(1) And equally dividing the residual target features into 10-20 boxes according to the quantiles, and calculating the median of the values of each box and the corresponding bad account rate.
(2) The change rate (slope) is calculated using the median of the values of each bin and the preceding bin and the corresponding bad account rate.
(3) And counting the number of the boxes which are larger than 0 and the non-0 boxes in the change rate between two adjacent boxes, and calculating the percentage of the number of the boxes which are larger than 0 in the change rate to the number of the non-0 boxes in the change rate.
(4) And acquiring the risk characteristic preset by the characteristic, and according to the percentage of the number of boxes larger than 0 in the calculated change rate to the number of boxes other than 0 in the change rate, judging whether the actual bad account rate change trend is approximately consistent with the business logic bad account rate change trend.
The criteria for approximate agreement are as follows:
First, if the business logic trend of this feature should be: as this feature value increases, the bad account rate increases (e.g., credit line usage, etc.). And in the module, eliminating the characteristics, of which the number of boxes which are larger than 0 in the calculated slope accounts for less than 70 percent of the non-0 boxes, namely eliminating the characteristics of which the characteristics show no business trend.
Second, if the business logic trend of this feature should be: as this feature value increases, the bad account rate decreases (e.g., deposit amount, etc.). And in the module, eliminating the characteristics, of which the number of boxes which are larger than 0 in the calculated slope accounts for more than 30 percent of the non-0 boxes, namely eliminating the characteristics of which the characteristics show no business trend.
90 Features were removed by the fifth round of screening, with 118 remaining features.
For the Beta6 sub-model divided in example 1, the variables selected are a customer base information class and a personal financial asset class variable (total of 316 feature variables) for approximately predicting the probability of future credit overdue for a registered customer without credit application, and the feature prescreening step and subsequent steps are the same as for the other sub-models.
Example 3 conversion of the features after Primary screening
And a feature judging step, wherein an optimal conversion mode is selected according to the concentration degree, the data type and the like of the features in the feature preliminary screening step. When judging the conversion mode, the data types are generally divided into two main types, one type is a character type variable and the other type is a numerical type variable. For character type variables, a conversion mode of dummy features (dummy variables) is generally adopted for variable conversion; for a numerical variable, a WOE conversion method is used if the variable value is less than about 5, and a WOE or continuous conversion method is used if the variable value is more (an optimal conversion method is selected by the correlation with the target variable). In the process, the condition of variable concentration is needed to be comprehensively considered, if continuous variable value is more, but the concentration of single value exceeds 95%, continuous processing is not carried out, and a WOE conversion mode is directly used. Features that employ different transformation schemes are grouped and partitioned into different data sets, and, in particular,
The remaining 118 features in embodiment 2 are first determined, and the following three methods are selected according to the concentration of the features, the data type, and the like.
And the feature conversion mode 1 is used for carrying out WOE conversion on the feature of the WOE conversion judged as the optimal conversion mode in the feature judgment module acquired from the data acquisition module.
And the feature conversion mode 2 is used for converting the features of the optimal conversion mode judged to be the dummy feature conversion in the feature judgment module acquired in the data acquisition module into the dummy feature conversion.
And the feature conversion mode 3 is used for selecting the optimal continuous conversion mode to perform continuous conversion by acquiring the features which are judged to be optimal to be continuous conversion in the feature judgment module in the data acquisition module.
And the feature merging module is used for transversely splicing the data of the feature conversion mode 1, the feature conversion mode 2 and the feature conversion mode 3.
Through this example, 118 features were transformed, 17 of them were WOE-transformed, 6 of them were dummy-transformed (amplified to 15 variables), and 95 of them were continuously transformed. The transformation of the as-screened features for the Beta6 sub-model is also substantially similar.
Example 4 feature depth screening (feature fine screening step)
In the embodiment, the following 4 steps are mainly performed, and the features with larger multiple collinearity in the feature merging module are removed based on stepwise regression and variance expansion factor calculation in the steps 1 and 2, so that the robustness of the model is enhanced. And 3, eliminating the feature that the signs of the training coefficients in the model do not accord with the business trend. Step 4 uses the population stability factor (PSI) to reject unstable features. The present embodiment uses LOGISTIC processes in SAS for depth screening.
And a feature fine screening step 1, which is based on a stepwise regression algorithm and on an F test and T test method, wherein features are sequentially introduced from high to low on the basis of significance, and each time a feature is introduced, the selected features are tested one by one. The features originally introduced are rejected when they become no longer significant due to the introduction of the following features. This process is cycled until no features above the significance threshold are selected into the equation, or no features below the significance threshold are removed from the regression equation. Through this step, 118 features are culled to 54 features.
And (2) a feature fine screening step, which is based on a method for eliminating features with higher variance expansion factors, and further reduces multiple collinearity in the model. Through this step, 54 features are culled to 48 features.
And 3, comparing the risk characteristics (preset by the system) of various risk points with the positive and negative signs of the model training coefficients, judging whether the characteristic coefficients of the residual characteristics in the model in the step 3 accord with the business trend, eliminating the characteristics of the model coefficients not accord with the business trend, and iterating again.
The specific embodiment of the feature fine screening module 3 is as follows:
1. For the characteristics of WOE class in the characteristic conversion mode, the corresponding model training coefficient should be negative, and the WOE conversion class characteristics of which the training coefficient is positive should be removed.
2. For the continuous conversion mode, if the feature value is increased in service logic, the bad account rate should be increased (such as credit line usage rate, etc.), the corresponding model training coefficient should be positive, and the continuous conversion type feature with the training coefficient being negative should be removed; if the feature value increases in service logic and the bad account rate should decrease (e.g. deposit amount, etc.), the corresponding model training coefficient should be negative, and the continuous conversion type feature with the training coefficient being positive should be removed.
3. For the dummy feature conversion mode, the bad account rate with the value of 1 and the value of 0 needs to be judged, if the bad account rate with the value of 1 is larger than the bad account rate with the value of 0, the coefficient is positive, otherwise, the coefficient is negative.
And a feature fine screening step 4, namely a data stability monitoring step, wherein the data stability monitoring step is used for evaluating whether obvious deviation occurs to the distribution situation of single features and integral scores of different points, the features of PSI >0.25 are directly removed, and the features of 0.25 PSI >0.1 are removed with cautions according to the influence of removing the features on the model distinguishing capability.
And (3) carrying out multiple iterations on the feature list based on the steps (3) and (4), stopping the iteration until no new feature is added or removed from the model finally, and obtaining a final feature list and a conversion value thereof. Through the steps, 48 features are removed as feature variables of the last modeling, for example, 5 features, 6 features, 7 features and 8 features can be adopted.
EXAMPLE 5 construction of Beta1 model
In general, the stronger the correlation between the characteristic variable and the target variable is, the more the accuracy of the final model can be improved. In this embodiment, beta1 is used as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 32 Mo Zuo, the guest group is mainly the current unexpired customer who has applied for credit service, the credit utilization rate is less than 100%, the current asset size is less than 500 yuan, and the model is constructed for predicting the probability of occurrence of overdue credit for 60 days or more for the crowd.
Final confirmation in this example 5 the final modeling results were described taking as an example six features, the current time point credit account balance, the last 12 months credit account maximum overdue number, the last 6 months credit account time point balance loop ratio maximum consecutive decrease month count, the current credit card remaining usable credit, the last 12 months wage maximum, the cycle credit last 6 months credit usage minimum. In the feature conversion step conversion mode selection, 6 modulo variables have been converted in a corresponding manner according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 6 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC process of SAS software) to perform logistic regression calculation to calculate a model of the credit violation probability.
The balance of the deposit account at the current time point represents the asset level of the borrower. For customers who are not overdue at present, have credit utilization rates of less than 100% and have fewer assets, the repayment capability of the borrower is intuitively reflected. The lower the balance of the deposit account at the current time point, the lower the repayment capability of the borrower, and the higher the possibility of serious default; conversely, the higher the balance of the deposit account at the current point in time, the greater the borrower's repayment capacity and the lower the likelihood of serious default. The conversion mode is natural logarithmic conversion. The conversion mode is natural logarithmic conversion in continuous conversion.
The maximum overdue number of the credit clients in the past 12 months represents the historical credit behavior of the borrower. For customers who are not currently overdue and have credit usage rates of less than 100% and less assets, the smaller the maximum number of overdue in the past 12 months, the lower the likelihood of serious violations; conversely, the greater the number of maximum expiration dates over the past 12 months, the worse the borrower's credit response, and the greater the likelihood of serious violations. The conversion mode is original value conversion in continuous conversion.
The balance ring ratio of the past 6 month deposit account time point is reduced continuously by the maximum month number ratio, so that the fluctuation of the asset change level of borrowers is reflected. For customers who are not overdue and have credit usage rates less than 100% and less assets, a higher continuous decrease in the time balance of the deposit account for a month means a longer continuous decrease in the asset size, and a higher probability of serious violations. Conversely, a lower point balance ring for the past 6 month deposit account than the maximum continuous reduction month count represents a relatively shorter continuous reduction time for the asset size, and represents a more stable borrower funds, with a reduced likelihood of serious violations. The conversion mode is a cube root conversion in continuous conversion.
The remaining available credit line of the current credit card reflects the credit requirement of the borrower. For customers who are not overdue at present, have credit utilization rates of less than 100% and have fewer assets, the lower the current credit card remaining available line, the greater the borrower's fund pressure, the lower the likelihood of serious violations; conversely, the higher the current credit card remaining available, the less funds are pressed by the borrower and the greater the likelihood of serious default. The conversion mode is a cube root conversion in continuous conversion.
The maximum payroll over the past 12 months represents the borrower income level. For customers who are not currently overdue and have credit usage rates of less than 100% and less assets, the lower the maximum payroll for the past 12 months, the lower the overall income level for the borrower, the higher the likelihood of serious violations; conversely, the higher the maximum payroll over the past 12 months, the higher the overall revenue level for the borrower and the lower the likelihood of serious violations. The conversion mode is dummy feature conversion.
The minimum value of the utilization rate of the line of the past 6 months of cyclic credit reflects the recent credit demand of borrowers. For customers who are not overdue at present, have credit usage rates of less than 100% and have more assets, the lower the minimum value of the credit line usage rate is in the past 6 months, the lower the repayment pressure of borrowers is, and the lower the possibility of serious default is; conversely, the higher the minimum value of the usage rate of the line of the past 6 months of the cyclical credit, the greater the borrower repayment pressure, and the greater the likelihood of serious violations. The conversion mode is square root conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
where k is the number of features of the model entered, and k is 6 in equation 1.
Alpha is an intercept term, and the numerical range is (0.20792, -0.57608), and most preferably-0.18408; β1 is the corresponding coefficient of the balance of the deposit account at the current time point, the numerical range is (-0.41409, -0.45329), and the most optimal value is-0.43369; β2 is the maximum overdue number corresponding coefficient for the past 12 months of credit customers, the range of values (-0.65896, -0.69816), and most preferably-0.67856; beta 3 is the corresponding coefficient of the maximum continuous reduction month number of the point balance ring ratio of the deposit account of the past 6 months, the numerical range is (-0.47066, -0.50986), and the most optimal value is-0.49026; beta 4 is the corresponding coefficient of the residual usable credit card, the numerical range is (-0.15071, -0.18991), and the most preferable value is-0.17031; beta 5 is the corresponding coefficient of the maximum payroll value in the past 12 months, the numerical range is (-1.08474, -1.16314), and the most preferable value is-1.12394; beta 6 is the corresponding coefficient of the minimum value of the limit usage rate of the cyclic credit for the past 6 months, the value range is (-0.51313, -0.70913), and the most preferable value is-0.61113. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is a natural logarithmic conversion value of the balance of the deposit account at the current time point generated in the feature conversion step; x2 is the original value of the maximum overdue number of the past 12 month credit customer generated by the feature conversion step; x3 is a cube root conversion value of the time point balance ring ratio of the past 6 month deposit account generated in the feature conversion step, which is the maximum continuous reduction month number ratio; x4 is the cube root conversion value of the residual usable credit of the current credit card generated in the feature conversion step; x5 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step; x6 is the square root transformed value of the minimum usage of the credit for the last 6 months of the cycle generated by the feature transformation step. The model of the partial features is presented in table 2 below.
TABLE 2
Beta 1
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
EXAMPLE 6 construction of Beta2 model
Aiming at Beta2 as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 180 ten thousand, the guest group is mainly a customer who has applied for credit service and is not overdue at present, the credit utilization rate is less than 100%, the current asset size is more than or equal to 500 yuan, and the model is constructed for predicting the probability (target variable) that the crowd of the guest group generates credit overdue for 60 days or more.
In this example 6, the final modeling result is described by taking as an example five features of the maximum consecutive months of the past 12-month line usage >70%, the minimum of the cyclic credit past 6-month line usage, the current credit card remaining usable line, the past 3-month monthly-credit-present-month refund amount/month-average asset size value, and the current time-point deposit account balance. In the feature conversion step conversion mode selection, 5 modulo variables have been converted in a corresponding form according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 5 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC processes of SAS software are utilized) to carry out logistic regression calculation on the model of the credit violation probability.
The maximum continuous month number of the past 12 months of the use rate of the forehead is more than 70 percent, which reflects the long-term credit requirement of borrowers. For customers who are not overdue at present, have credit utilization rates of less than 100% and have more assets, the greater the maximum number of consecutive months for which the use rate of the line of the past 12 months is greater than 70%, the greater the repayment pressure of the customer, the higher the probability of serious default; conversely, the less the maximum number of consecutive months the rate of use of the forehead over the past 12 months >70%, the less likely serious violations will occur. The conversion mode is original value conversion in continuous conversion.
The minimum value of the utilization rate of the line of the past 6 months of cyclic credit reflects the recent credit demand of borrowers. For customers who are not overdue at present, have credit usage rates of less than 100% and have more assets, the lower the minimum value of the credit line usage rate is in the past 6 months, the lower the repayment pressure of borrowers is, and the lower the possibility of serious default is; conversely, the higher the minimum value of the usage rate of the line of the past 6 months of the cyclical credit, the greater the borrower repayment pressure, and the greater the likelihood of serious violations. The conversion mode is square root conversion in continuous conversion.
The remaining available credit line of the current credit card reflects the credit requirement of the borrower. For customers who are not overdue at present, have credit utilization rates of less than 100% and have more assets, the lower the residual available credit card amount is, the higher the fund pressure of borrowers is, and the lower the possibility of serious default is; conversely, the higher the current credit card remaining available, the less funds are pressed by the borrower and the greater the likelihood of serious default. The conversion mode is natural logarithmic conversion in continuous conversion.
The lowest payoff amount per month average asset size of the past 3 month average credit month reflects the credit repayment capability of borrowers. For customers who are not currently overdue and have credit usage rates of less than 100% and more assets. When the lowest payoff amount per month average asset size of the past 3 months average credit month is larger than 1, the higher the debt repayment ratio of the borrower is, the higher the liability of the repayment pressure is, the higher the liability is, and the higher the possibility of serious default is; conversely, when the minimum payoff amount per month of average asset size for the past 3 months of average credit is less than 1, the borrower payoff pressure is lower the asset is higher than liability, and the probability of serious default is reduced. The conversion mode is a cube root conversion in continuous conversion.
The balance of the deposit account at the current time point reflects the deposit level of the borrower. For customers who are not currently overdue and have credit usage rates of less than 100% and more assets, current credit is an important indicator of recent repayment capability for areas of moderate economic development. The lower the balance of the deposit account at the current time point, the weaker the borrower's recent debt repayment ability, and the higher the probability of serious default; conversely, the higher the balance of the deposit account at the current point in time, the greater the borrower's recent repayment capacity and the reduced the likelihood of serious default. The conversion mode is natural logarithmic conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
where k is the number of features of the model entered, and k is 5 in equation 2.
Alpha is the intercept term, the numerical range is (-0.27492, -0.66692), and most preferably-0.47092; β1 is the maximum continuous month number corresponding coefficient with the use rate of the forehead of the past 12 months being more than 70%, the numerical range is (0.19904-0.03616), and the most preferable value is 0.081443; beta 2 is the minimum value corresponding coefficient of the limit usage rate of the cyclic credit for the past 6 months, the numerical range is (0.106, -0.09), and the optimal value is 0.008002; beta 3 is the corresponding coefficient of the residual usable credit card, the numerical range is (-0.49447, -0.76887), and the most preferable value is-0.63167; beta 4 is the corresponding coefficient of the amount to be refund per month of the past 3 months monthly credits per month average asset scale value, the numerical range is (0.24108, -0.22932), and the optimal value is 0.005877; beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the numerical range is (-0.6779, -0.9131), and the most preferable value is-0.7955. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is the original value of the maximum consecutive months of which the rate of use of the line of the past 12 months generated in the feature conversion step is > 70%; x2 is the square conversion value of the minimum value of the use rate of the line of the past 6 months of the cyclic credit generated in the characteristic conversion step; x3 is the natural logarithmic conversion value of the current credit card remaining usable credit generated in the feature conversion step; x4 is the cube root conversion value of the asset scale of the amount to be refund/month average for the last 3 months monthly credits generated by the feature conversion step; x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step. The model of the partial features is presented in table 3 below:
TABLE 3 Table 3
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
EXAMPLE 7 construction of Beta3 model
Aiming at Beta3 as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 6 ten thousand, the guest group is mainly a customer who has applied for credit service and is not overdue at present, the credit utilization rate is more than or equal to 100%, the current asset size is less than 500 yuan, and the model is constructed for predicting the probability (target variable) that the crowd generates credit overdue for 60 days or more.
In this example 7, the final modeling result was described by taking six features of the minimum value of the usage rate of the line of the past 6 months of the cyclic credit, the usage rate of the line of the past 12 months of the cyclic credit >10% of the month count ratio, the usage rate of the line of the past 12 months of the credit >70% of the month count ratio, the average asset size of the line of the past 6 months/the average asset size of the past 12 months, the usage rate of the line of the past 3 months >90% of the month count ratio, and the minimum balance at the time of the deposit account of the past 3 months as an example. In the feature conversion step conversion mode selection, 6 modulo variables have been converted in a corresponding manner according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 6 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC process of SAS software) to perform logistic regression calculation to calculate a model of the credit violation probability.
The minimum value of the utilization rate of the line of the past 6 months of cyclic credit reflects the recent credit demand of borrowers. For customers who are not overdue and have credit usage exceeding 100% and less assets, the lower the minimum of the last 6 months of cyclic credit line usage, the lower the borrower repayment pressure, the lower the likelihood of serious default; conversely, the higher the minimum value of the usage rate of the line of the past 6 months of the cyclical credit, the greater the borrower repayment pressure, and the greater the likelihood of serious violations. The conversion mode is a cube root conversion in continuous conversion.
The cycle credit has a use ratio of more than 10% for the past 12 months of the line, which reflects the credit requirement and the use habit of borrowers. For customers who are not overdue and have credit usage exceeding 100% and less assets, credit loans (including credit cards) are generally high in credit line, so that the cyclic line usage represents the credit of the customers. The lower the month number proportion of the cycle credit for which the use ratio of the line of the past 12 months is more than 10%, the lower the credit demand of the borrower, the lower the probability of serious default is; conversely, the higher the month rate of the cycle credit over 12 months of use >10%, the higher the credit demand on the borrower, and the greater the likelihood of serious default. The conversion mode is original value conversion in continuous conversion.
The past 12 months of credit use rate is more than 70% of months, which reflects the long-term credit requirement of borrowers. For customers who are not currently overdue and have credit usage exceeding 100% and less assets, the greater the number of months for which the past 12 months of credit line usage is greater than 70%, the greater the customer's repayment pressure, the greater the likelihood of serious violations; conversely, the fewer months the past 12 months of credit line usage is greater than 70%, the less likelihood of serious violations will occur. The conversion mode is original value conversion in continuous conversion.
The asset size of the past 6 months/the asset size of the past 12 months reflects the condition of the asset level change of borrowers. For customers who are currently unexpired and have credit usage exceeding 100% and fewer assets, an increase in their asset level means an ever-increasing asset size. The greater the ratio of the past 6 month average asset size to the past 12 month average asset size is greater than 1, the more stable the borrower's recent asset level is, and the likelihood of an breach will be reduced. The worse the recent asset level of the borrower the less stable the fund level the less the ratio of the past 6 month average asset size/the past 12 month average asset size is less than 1, the higher the likelihood of an breach. The conversion mode is a cube root conversion in continuous conversion.
The past 3 months of the use rate of the forehead is more than 90% of the number of months, which reflects the recent credit requirement of borrowers. For customers who are not currently overdue and have credit usage exceeding 100% and less assets, the greater the number of months for which the past 3 months of credit line usage is greater than 90%, the greater the customer's repayment pressure, the greater the likelihood of serious violations; conversely, the less months the past 3 months of credit line usage is greater than 90%, the less likelihood of serious violations will occur. The conversion mode is dummy feature conversion.
The minimum balance of the past 3 months deposit account time point represents the deposit condition of the recent borrower. For customers who are not overdue and have credit usage exceeding 100% and less assets, the minimum deposit is more indicative of their asset quality. The higher the minimum balance of the deposit account over the past 3 months, the more abundant the borrower funds are, and the lower the probability of serious violations is; conversely, the lower the minimum balance of the deposit account over the last 3 months, the less funds the borrower has, and the greater the likelihood of serious violations. The conversion mode is natural logarithmic conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
Where k is the number of features of the model entered, and k is 6 in equation 3.
Alpha is the intercept term, the numerical range is (-0.23953, -0.27873), and most preferably-0.25913; β1 is the minimum corresponding coefficient of the usage rate of the line of the last 6 months of the cyclic credit, the numerical range is (-0.4522, -0.4914), and the most preferable value is-0.4718; beta 2 is the corresponding coefficient of month number proportion of which the use rate of the line of the past 12 months of the cyclic credit is more than 10 percent, the numerical range is (-0.22933, -0.26853), and the most preferable value is-0.24893; beta 3 is the month number corresponding coefficient of which the credit limit use rate is more than 70% in the past 12 months, the numerical range is (-0.51155, -0.58995), and the most preferable range is-0.55075; beta 4 is the corresponding coefficient of the average asset size of the past 6 months/the average asset size of the past 12 months, the numerical range is (-0.19469, -0.27309), and the most preferable is-0.23389; beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, the numerical range is (-0.22323, -0.26243), and the most preferable range is-0.24283; beta 6 is the minimum balance corresponding coefficient of the past 3 month deposit account time point, the numerical range is (-0.80785, -0.84705), and the most preferable value is-0.82745. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is a cube root conversion value of the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit generated in the characteristic conversion step; x2 is the original value of the month number occupation ratio of the cyclic credit generated in the feature conversion step, wherein the use rate of the line of the past 12 months is more than 10%; x3 is the original value of the number of months for which the credit generated in the feature conversion step has a usage rate of >70% for the past 12 months; x4 is the cube root conversion value of the last 6 month average asset size/the last 12 month average asset size generated in the feature conversion step; x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step; x6 is the natural log-transformed value of the minimum balance of the past 3 month deposit account timepoints generated by the feature transformation step.
The model of the partial features is presented in table 4 below:
TABLE 4 Table 4
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
Example 8Beta4 model construction
Aiming at Beta4 as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 11 ten thousand, the guest group is mainly a customer who has applied for credit business and is not overdue at present, the credit utilization rate is more than or equal to 100%, the current asset size is more than or equal to 500 yuan, and the model is constructed for predicting the probability (target variable) that the crowd of the guest group generates credit overdue for 60 days or more.
In example 8, final modeling results were described taking as an example seven features of the average asset size of the past 12 months, the minimum balance of the past 3 month financial account from the number of months of the present, the average asset size of the past 3 months monthly credit that should be paid back/month, the minimum balance of the past 6 month deposit account time point, the number of months of the past 3 month line usage >90%, whether to send out payouts, the past 6 month financial product balance/the past 12 month financial product balance. In the feature conversion step conversion mode selection, 7 modulo variables have been converted in a corresponding form according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 7 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC process using SAS software) to perform logistic regression to calculate a model of the credit violation probability.
Average asset size over the past 12 months represents the borrower's repayment capability. For customers who are not currently overdue and have credit usage exceeding 100% and more assets, the lower the average asset size over the past 12 months, the lower the borrower's repayment capacity, and the higher the likelihood of serious violations; conversely, the higher the average asset size over the past 12 months, the greater the borrower's repayment capacity and the lower the likelihood of serious violations. The conversion mode is natural logarithmic conversion in continuous conversion.
The minimum balance of the financial account in the past 3 months is the number of the present months, which reflects the recent financial asset change of borrowers. For customers who are not overdue at present, have credit utilization rate exceeding 100% and have more assets, the higher the minimum balance of the financial account of the past 3 months is from the current month, the higher the recent financial assets are, and the lower the possibility of serious default is; conversely, serious violations are more likely to occur. The conversion mode is original value conversion in continuous conversion.
The last 3 months monthly of credit should be paid per month of the average asset size, which reflects the credit repayment capability of borrowers. For customers who are not overdue at present and have credit usage rate exceeding 100% and more assets, when the average minimum amount of money to be paid per month of the past 3 months is larger than 1 on a large scale, the higher the debt repayment ratio of borrowers is, the larger the debt repayment pressure is, the higher the liability is, and the higher the possibility of serious default is; conversely, when the minimum payoff amount per month of average asset size for the past 3 months of average credit is less than 1, the borrower payoff pressure is lower the asset is higher than liability, and the probability of serious default is reduced. The conversion mode is a cube root conversion in continuous conversion.
The minimum balance of the past 6 month deposit account time points reflect the deposit condition of borrowers. For customers who are not overdue and have credit usage exceeding 100% and more assets, the minimum balance of the deposit account can better represent the asset quality. The higher the minimum balance of the deposit account at the past 6 months, the more abundant the borrower funds are, and the lower the probability of serious default is; conversely, the lower the minimum balance at the time of the past 6 month deposit account, the less funds the borrower has, and the greater the likelihood of serious violations. The conversion mode is natural logarithmic conversion in continuous conversion.
The past 3 months of the use rate of the forehead is more than 90% of the number of months, which reflects the recent credit requirement of borrowers. For customers who are not currently overdue and have credit usage exceeding 100% and more assets, this variable reflects their repayment pressure. The more months the usage of the line of credit over the past 3 months is greater than 90%, the greater the customer's repayment pressure, the greater the likelihood of serious violations; conversely, the less months the past 3 months of credit line usage is greater than 90%, the less likelihood of serious violations will occur. The conversion mode is dummy feature conversion.
Whether the borrower is a substitute payroll client or not reflects the stability of the current fund source of the borrower. For customers who are not overdue and have credit usage exceeding 100% and more assets, if they are payroll customers, the repayment pressure of the customers can be measured according to the income situation. The lower the probability of serious default is, the lower the probability is for the payroll client to generate due to the stable fund source; conversely, non-shipping payroll customers may be unable to monitor due to the source of funds and the likelihood of serious violations may increase. The conversion mode is dummy feature conversion.
The balance of the financial product in the past 6 months/the balance of the financial product in the past 12 months reflects the property change condition of borrowers. For customers who are not overdue at present, have credit utilization rate exceeding 100% and have more assets, the larger the ratio of the balance of the financial product in the past 6 months to the balance of the financial product in the past 12 months, the higher the recent financial asset of the borrower, and the lower the possibility of serious default; conversely, serious violations are more likely to occur. The conversion mode is square root conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
Where k is the number of features of the model entered, and in equation 4, k is 7.
Alpha is the intercept term, the numerical range is (3.5122,3.4338), and most preferably 3.473; β1 is the corresponding coefficient of average asset size over the past 12 months, the range of values (-0.1731, -0.2123), optimally-0.1927; beta 2 is the corresponding coefficient of the minimum balance of the financial account of the past 3 months to the number of months, the numerical range is (-0.21993, -0.29833), and the most preferable value is-0.25913; beta 3 is the corresponding coefficient of the size of the repayment/month-average asset of the past 3 months monthly credit, the numerical range is (-0.22933, -0.26853), and the optimal value is-0.24893; beta 4 is the minimum balance corresponding coefficient of the past 6 month deposit account time point, the numerical range is (-0.53115, -0.57035), and the most preferable value is-0.55075; beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, the numerical range is (-0.27928, -0.35768), and the most preferable range is-0.31848; β6 is a corresponding coefficient of whether to replace the wage, the numerical range is (-0.21913, -0.25833), and the most preferable range is-0.23873; beta 7 is the corresponding coefficient of the balance of the financial product in the past 6 months/the balance of the financial product in the past 12 months, and the numerical range is (-0.55257, -0.63097), and is most preferably-0.59177. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is the natural logarithmic conversion value of the average asset size over the past 12 months generated by the feature conversion step; x2 is the original value of the minimum balance of the financial account of the past 3 months generated in the feature conversion step from the number of months; x3 is the cube root conversion value of the monthly payment/monthly average asset size of the past 3 months monthly credit generated in the feature conversion step; x4 is a natural logarithmic conversion value of the minimum balance of the past 6 month deposit account time points generated in the feature conversion step; x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step; x6 is a dummy feature conversion value of whether to replace wages generated in the feature conversion step; x7 is the square root conversion value of the past 6 month financial product balance/the past 12 month financial product balance generated by the feature conversion step.
The model of the partial features is shown in table 5 below:
TABLE 5
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
EXAMPLE 9Beta5 model construction
Aiming at Beta5 as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 15 ten thousand, the guest group is mainly the current overdue client applying for credit service, and the model is constructed for predicting the probability (target variable) that the crowd of the model is overdue for 60 days or more.
In example 9, the final modeling results are described by taking five characteristics of the number of months of cyclic credit for which the usage rate of the line of the last 6 months is >70%, the average of the overdue balances of the credit customers for the last 3 months, the percentage of the overdue balances of the credit customers for the last 6 months to the loan amount, the current account age of the credit customers, and the maximum value of the repayment amount for the last 3 months of the credit as examples. In the feature conversion step conversion mode selection, 5 modulo variables have been converted in a corresponding form according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 5 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC processes of SAS software are utilized) to carry out logistic regression calculation on the model of the credit violation probability.
The cyclic credit has a use ratio of the line of the past 6 months of more than 70% of the number of months, which reflects the long-term credit demand of borrowers. For overdue customers, the greater the number of months the past 6 month cycle credit has been used with a rate greater than 70%, the greater the customer's repayment pressure, and the greater the likelihood of serious default; conversely, the fewer months the cycle credit usage of greater than 70% over the past 6 months, the less likelihood of serious violations will occur. The conversion mode is original value conversion in continuous conversion.
The average value of the overdue balances of the credit clients in the past 3 months reflects the recent overdue condition of borrowers. For overdue clients, the higher the average value of overdue balances of the current clients for 3 months, the higher the overdue degree of borrowers, and the higher the continuous default possibility; conversely, the lower the average of the over-term balances of the current customer's 3-month credit customers, the weaker the borrower's over-term, and the higher the likelihood of exiting the default condition. The conversion mode is natural logarithmic conversion in continuous conversion.
The overdue balance of the past 6 months credit client accounts for the percentage of the loan amount, and represents the overdue condition of the long-term middle period of the borrower. For overdue customers, the higher the percentage of overdue balances of the past 6 months credit customers to the loan amount, the higher the overdue degree of borrowers, the higher the likelihood of continuous default; conversely, the lower the percentage of the past 6 months credit customer's overdue balance in the amount of the loan, the weaker the borrower's overdue, and the higher the likelihood of exiting the default condition. The conversion mode is dummy feature conversion.
The current account age of the credit client reflects the current credit history length of the borrower, and the longer age is, the stronger the credit product management capability is, the less possibility of bad account occurrence is for the overdue client. Conversely, the shorter age, the weaker the credit product management capability, and the higher the likelihood of default. The conversion mode is WOE conversion.
The maximum value of repayment amount of 3 months in the past of credit represents the recent repayment capability of borrowers. For overdue customers, the maximum payoff amount is helpful in measuring the borrower's asset level and payoff ability. The higher the maximum value of the refund amount in the past 3 months, the greater the refund ability of the borrower and the lower the likelihood of continuous default; conversely, the lower the maximum value of the refund amount over the past 12 months, the lower the borrower's recent refund capacity and the greater the likelihood of default. The conversion mode is square root conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
where k is the number of features of the model entered, and in equation 5, k is 5.
Alpha is the intercept term, the numerical range is (2.9266,2.7698), and most preferably 2.8482; β1 is a month number corresponding coefficient with a cyclic credit usage rate of >70% for the past 6 months, the numerical range is (0.05201, -0.06559), and most preferably-0.00679; β2 is the average corresponding coefficient of the overdue balances of the credit customers of the past 3 months, the numerical range is (0.03734, -0.04106), and the most preferable value is-0.00186; beta 3 is the corresponding coefficient of the percentage of the overdue balance of the past 6 months credit customer to the loan amount, the numerical range is (-0.10043, -0.21803), and the most preferable is-0.15923; beta 4 is the corresponding coefficient of the current account age of the credit client, the numerical range is (0.03485, -0.20035), and the optimal value is-0.08275; beta 5 is the maximum value corresponding coefficient of the repayment amount of the past 3 months of credit, and the value range is (-0.36818, -0.44658), and is most preferably-0.40738. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is the original value of the number of months that the cyclic credit generated in the feature conversion step has a usage rate of the line of the past 6 months of > 70%; x2 is a natural logarithmic conversion value of the average value of the overdue balances of the past 3-month credit customers generated in the feature conversion step; x3 is a dummy feature conversion value of the percentage of the overdue balance of the past 6 month credit customer to the loan amount generated in the feature conversion step; x4 is the WOE conversion value of the current account age of the credit client generated in the feature conversion step; x5 is the square root conversion value of the maximum value of the repayment amount of the credit for the past 3 months generated by the feature conversion step.
The model of the partial features is presented in table 6 below:
TABLE 6
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
Example 10Beta6 model construction
Aiming at Beta6 as a sub-model of the subdivision model, the decision tree method is adopted, the sample size obtained by classification is about 3 ten thousand, the guest group is mainly a customer with account age smaller than 3 months of applied credit service, and the model is constructed for predicting the probability (target variable) that the future occurrence credit overdue of the sample group without applied registered credit service is 60 days or more.
In example 10, final confirmation the final modeling results are described taking five features of the past 6 months of gold account holding month number, the past 12 months of investment financial account average balance, the past 12 months of wage maximum value, the past 3 months of asset size/past 12 months of asset size average value, the current time point deposit account balance as examples. In the feature conversion step conversion mode selection, 5 modulo variables have been converted in a corresponding form according to the correlation of the feature and the target variable. In the credit violation probability modeling step, 5 features screened in the feature fine screening step are substituted into a Sigmoid function (LOGISTIC processes of SAS software are utilized) to carry out logistic regression calculation on the model of the credit violation probability.
The past 6 months of the gold account holds the month number, which reflects the credit worthiness of borrowers. For the new application group, the higher the number of months the gold account has in the past 6 months, the higher the borrower's reputation, the lower the likelihood of serious violations; conversely, the lower the number of months the gold account has been held for the past 6 months, the lower the borrower's reputation, and the greater the likelihood of serious violations. The conversion mode is original value conversion in continuous conversion.
The average balance of the financial account is invested in the past 12 months, so that the credit condition of borrowers is reflected. For the new application group, the higher the number of months the financial account holds in the past 12 months, the higher the credit of the borrower, and the lower the probability of serious default; conversely, the lower the number of months the financial account holds over the past 12 months, the lower the borrower's reputation, and the greater the likelihood of serious violations. The conversion mode is natural logarithmic conversion in continuous conversion.
The maximum payroll over the past 12 months represents the borrower income level. The lower the maximum payroll over the past 12 months, the lower the overall income level of the borrower and the higher the likelihood of serious violations; conversely, the higher the maximum payroll over the past 12 months, the higher the overall revenue level for the borrower and the lower the likelihood of serious violations. The conversion mode is a dummy feature conversion mode.
The average value of the past 3 month asset scale/the past 12 month asset scale represents the asset level change condition of borrowers. For new applicant groups, the increase in asset levels means that the asset size is continually increasing. The higher the average value ratio of the average asset size of the past 3 months to the average asset size of the past 12 months is greater than 1, the more stable the borrower funds are, the better the recent funds are, and the probability of occurrence of the default is reduced. The lower the average of the past 3 month average asset size/the past 12 month average asset size is less than 1, the worse the borrower's recent funds status, the more unstable the funds level, and the higher the likelihood of occurrence of a breach. The conversion mode is a cube root conversion in continuous conversion.
The balance of the deposit account at the current time point reflects the asset level of the borrower. For the newly applied guest group, the lower the balance of the deposit account at the current time point is, the lower the repayment capability of the borrower is, and the higher the possibility of serious default is; conversely, the higher the balance of the deposit account at the current point in time, the greater the borrower's repayment capacity and the lower the likelihood of serious default. The conversion mode is natural logarithmic conversion in continuous conversion.
In the step of calculating credit-default probability, the probability (P) of characterizing the borrower default is predicted using the following model formula according to the data and coefficient conditions acquired in the feature conversion step:
Where k is the number of features of the model entered, and in equation 6, k is 5.
Alpha is the intercept term, the numerical range is (-3.0574, -3.2926), and most preferably-3.175; β1 is the corresponding coefficient of the number of months the gold account holds over the past 6 months, the numerical range is (-0.76852, -0.96452), and most preferably-0.86652; β2 is the corresponding coefficient of the average balance of the investment financial account of the past 12 months, the numerical range is (-0.37245, -0.64685), and the most preferable value is-0.50965; beta 3 is the corresponding coefficient of the maximum payroll value in the past 12 months, the numerical range is (-0.81714, -1.28754), and the most preferable value is-1.05234; beta 4 is the average corresponding coefficient of the past 3 month asset scale/the past 12 month asset scale, the numerical range is (-0.0053, -0.2405), and most preferably-0.1229; beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the numerical range is (-0.92484, -1.04244), and the most preferable value is-0.98364. ( And (3) injection: the value range is from the 95% confidence interval, 95% CI in the table below )
X1 is the original value of the number of months held by the gold account for the last 6 months generated in the feature conversion step; x2 is a natural logarithm conversion value of the average balance of the investment financial account of the past 12 months generated in the characteristic conversion step; x3 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step; x4 is the cube root transformed value of the past 3 month asset size/past 12 month asset size average generated by the feature transformation step; x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step.
The model of the partial features is presented in table 7 below:
TABLE 7
The P-values for all model features were less than 0.05, indicating that the features were significantly correlated with default performance.
Example 11
The P value calculated for each of the above formulas may be further used to calculate a score for any one customer.
And a scoring step of converting the calculated credit violation probability into a score of 0-1000 points by pre-storing a default score conversion code.
In the calculate scoring module, a credit score for characterizing the borrower is calculated using the following formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
The P value calculated in each of the above embodiments may be substituted in the present embodiment to calculate the score.
KS (Kolmogorov-Smirnov) statistics were presented by two Soviet math A.N.Kolmogorov and N.V.Smirnov. In wind control, KS is commonly used to evaluate model discrimination. The larger the discrimination, the more risk ranking capability (ranking ability) of the model is explained.
KS statistics are established based on an empirical cumulative distribution function (EMPIRICAL CUMULATIVE DISTRIBUTION FUNCTION, ECDF), defined generally as:
KS=max(|cum(bad_rate)-cum(good_rate)|)
According to the prediction model, a KS curve is drawn, and KS of the overall Beta model is 74.96 in this embodiment, it can be seen that the model constructed in this embodiment performs well in evaluating the credit risk function of the client, and as a result, fig. 2 shows an example of a KS curve with a KS value of 58.
The following shows the results of the model applied to the development sample and the reserved validation sample for KS values in the development sample and the validation sample. It can be seen from table 8 that the model has a good distinguishing effect in the training set and the verification set (in which the sample number ratio of the training set to the verification set is 6:4) and the whole sample, i.e., the distinguishing effect is very excellent in all the sub-models.
TABLE 8
/>
Claims (56)
1. A method of calculating retail credit risk, comprising:
A data acquisition step of acquiring retail credit prediction data of a sample to be predicted;
Classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating credit violation probability;
and a credit violation probability calculating step, wherein retail credit prediction data are substituted into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted, and the credit violation probability is preferably the credit violation probability.
2. The method of claim 1, further comprising:
after calculating the credit violation probability, calculating a credit score of the sample to be predicted, for calibrating the calculated credit violation probability to a normalized score of 0-1000 points.
3. The method according to claim 1 or 2, wherein,
The retail credit prediction data comprises original retail credit prediction data of a sample to be predicted and derived retail credit prediction data is processed based on the original retail credit prediction data;
preferably, the original retail credit prediction data includes:
credit card based data, which is based on all available data in the sample user's credit card creation process and use process,
Personal loan class base data, which is all available data based on the sample user's loan application case and usage behaviors,
Customer base information class base data, which is data based on the properties of the sample user itself, but not directly related to behavior at the financial institution,
Personal financial property type base data, which is sample user data of all other financial properties and financial transaction types not related to credit cards and loans at a financial institution.
4. The method according to claim 1 to 3, wherein,
Processing derived retail credit prediction data based on the original retail credit prediction data refers to data obtained by processing the collected original retail credit prediction data based on a time dimension, a space dimension, a frequency dimension, and a statistical information dimension;
preferably, the derived retail credit prediction data includes, but is not limited to:
derived retail credit prediction data processed based on sample relationship lengths,
Derived retail credit forecast data processed based on time interval class variables,
Derived retail credit forecast data processed based on the degree of sample behavioral frequency,
Derived retail credit prediction data processed based on the sample current point in time conditions,
Derived retail credit forecast data processed based on sample duration behavior,
Derived retail credit prediction data processed from the sample data based on the statistical information dimension.
5. The method according to any one of claims 1 to 4, wherein,
The retail credit prediction data is selected from one or two or three or four or five or six or seven of the following:
The current time point credit account balance, the maximum expiration date of the past 12 month credit customer, the time point balance loop ratio of the past 6 month credit account, the maximum continuous reduction month rate, the current credit card remaining available amount, the past 12 month wage maximum, the cyclic credit past 6 month line usage minimum, the past 12 month line usage maximum continuous month rate >70%, the past 3 month monthly credit current month refund amount/month average asset size value, the cyclic credit past 12 month line usage >10% month rate, the credit past 12 month line usage >70% month number, the past 6 month average asset size/past 12 month average asset size, the past 3 month line usage >90% month number, the past 3 month credit account time point minimum balance, the cyclic credit account usage > 3% month average asset size/past 12 month average asset size value average property size of past 12 months, minimum balance of past 3 month financial account to date month, minimum balance of past 3 months monthly credit due month payoff/month average property size, past 6 month deposit account time point minimum balance, whether to send out wages, past 6 month financial product balance/past 12 month financial product balance, month number of cyclic credit past 6 month line usage >70%, past 3 month credit customer overdue balance average, percentage of past 6 month credit customer overdue balance to loan amount, credit customer current account age, credit past 3 month repayment amount maximum value, past 6 month gold account holding month number, past 12 month investment financial account average balance, past 3 month asset size/past 12 month asset size average.
6. The method according to any one of claims 1 to 5, wherein,
The step of classifying the samples to be predicted comprises the sub-steps of:
Whether the sample to be tested is a customer who has applied for credit services at a financial institution;
Whether the sample to be tested is a customer who has a credit overdue;
whether the current month credit utilization rate of the sample to be tested is more than or equal to a percent;
whether the current month asset scale of the sample to be tested is more than or equal to b;
Classifying samples to be predicted based on the substeps to determine a submodel for calculating credit violation probability, wherein the sequence of the substeps can be arbitrarily set on the premise of ensuring reasonable business logic;
The samples to be predicted are preferably classified in the following order:
firstly, judging whether the sample to be tested is a customer who applies for credit business at a financial institution;
Then judging whether the sample to be tested is a customer with overdue credit;
then judging whether the current month credit utilization rate of the sample to be tested is more than or equal to a percent;
And then judging whether the current month asset size of the sample to be tested is more than or equal to b.
7. The method according to any one of claims 1 to 6, wherein,
And substituting the feature conversion of the retail credit prediction data into a credit violation probability submodel to calculate the credit violation probability of the sample to be predicted, wherein the feature conversion step comprises the following steps:
And selecting a WOE mode, a dummy feature mode or a continuous mode to perform feature conversion based on the feature type of the retail credit prediction data which is required to be substituted into the credit violation probability submodel.
8. The method of claim 7, wherein,
The feature conversion in the continuous mode comprises the following steps: the continuous feature conversion is carried out in a mode of directly selecting an original value, calculating the square of original data, calculating the square root of the original data, calculating the cube root of the original data or calculating the natural logarithm of the original data.
9. The method according to any one of claims 1 to 8, wherein,
The credit violation probability submodel is a model constructed based on existing user populations using logistic regression based on sample retail credit prediction data and credit violation probabilities.
10. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the current time point credit account balance, the maximum overdue number of the past 12 month credit client, the time point balance loop ratio of the past 6 month credit account, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the past 12 month wage maximum value, the cyclic credit past 6 month amount use ratio minimum value of one, two, three, four, five or six.
11. The method of claim 10, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on the current time point deposit account balance, the maximum overdue number of the past 12 month credit clients, the time point balance loop ratio of the past 6 month deposit account, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the maximum wage value of the past 12 months and the minimum cyclic credit past 6 month amount use rate,
Preferably, converting the balance of the deposit account at the current time point in a continuous mode; converting the maximum overdue number of the past 12-month credit customer in a continuous mode; converting the point balance ring ratio of the past 6 month deposit account with the maximum continuous reduction month number ratio in a continuous mode; converting the residual usable credit of the current credit card in a continuous mode; converting the maximum wage value of the past 12 months in a dummy feature mode; converting the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit in a continuous mode;
Further preferably, the method comprises converting the current time point deposit account balance into a calculation method using natural logarithm of the current time point deposit account balance by adopting a continuous mode; converting the maximum overdue number of the past 12-month credit customer into a calculation mode using an original value for the maximum overdue number of the past 12-month credit customer in a continuous mode; converting the time point balance ring ratio of the past 6 month deposit account into a calculation mode of taking a cube root by adopting a continuous mode according to the time point balance ring ratio of the past 6 month deposit account; converting the current credit card residual usable amount into a calculation mode of taking a cube root for the current credit card residual usable amount; the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit is converted into a calculation mode of taking the square root of the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit by adopting a continuous mode.
12. The method of claim 11, wherein,
Substituting the values converted by the six characteristics of the current time point deposit account balance, the maximum overdue number of the past 12 month credit clients, the time point balance loop ratio of the past 6 month deposit accounts, the maximum continuous reduction month number ratio, the current credit card remaining usable amount, the wage maximum value of the past 12 months and the cyclic credit past 6 month amount use ratio minimum value into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
13. The method of claim 12, wherein,
Wherein the submodel is shown in formula 1 below:
where k is the number of features of the model entered, preferably k is 6,
Alpha is an intercept term, preferably in the range of values (0.20792, -0.57608), most preferably-0.18408;
β1 is the corresponding coefficient of the balance of the deposit account at the current time point, and the preferred value range is (-0.41409, -0.45329), and the most preferred value range is-0.43369;
β2 is the maximum expiration number corresponding to the last 12 months of credit customers, preferably ranging in value from (-0.65896, -0.69816), most preferably-0.67856;
β3 is the corresponding coefficient of the point balance ring ratio of the past 6 month deposit account with the maximum continuous reduction month number, and the preferable value range is (-0.47066, -0.50986), and the most preferable value range is-0.49026;
Beta 4 is the corresponding coefficient of the residual usable credit of the current credit card, and the preferred value range is (-0.15071, -0.18991), and most preferably-0.17031;
beta 5 is the maximum payroll corresponding coefficient for the past 12 months, the preferred range of values is (-1.08474, -1.16314), and most preferably-1.12394;
Beta 6 is the corresponding coefficient of the minimum value of the limit use rate of the cyclic credit for the past 6 months, the value range is (-0.51313, -0.70913), and the most optimal value is-0.61113;
x1 is a natural logarithmic conversion value of the balance of the deposit account at the current time point generated in the feature conversion step;
x2 is the original value of the maximum overdue number of the past 12 month credit customer generated by the feature conversion step;
x3 is a cube root conversion value of the time point balance ring ratio of the past 6 month deposit account generated in the feature conversion step, which is the maximum continuous reduction month number ratio;
x4 is the cube root conversion value of the residual usable credit of the current credit card generated in the feature conversion step;
x5 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step;
x6 is the square root transformed value of the minimum usage of the credit for the last 6 months of the cycle generated by the feature transformation step.
14. The method of claim 13, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
15. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: one, two, three, four or five of a maximum number of consecutive months of past 12 month line usage >70%, a minimum of past 6 month line usage of recurring credits, a current credit card remaining available line, a past 3 month monthly credit this month should be refunded/month average asset size value, a current point-in-time deposit account balance.
16. The method of claim 15, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on the maximum continuous month number of which the use rate of the credit line of the past 12 months is more than 70%, the use rate of the credit line of the past 6 months of the cyclic credit is the minimum value, the current credit card remaining usable amount, the amount to be paid per month of the past 3 months monthly credit, the asset scale value of the current time point deposit account balance,
Preferably, converting the maximum continuous month number of which the use rate of the forehead of the past 12 months is more than 70% in a continuous mode; converting the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit in a continuous mode; converting the residual usable credit of the current credit card in a continuous mode; converting the value of the amount to be refund per month of the past 3 months monthly credits per month of the average asset scale in a continuous mode; converting the balance of the deposit account at the current time point in a continuous mode;
Further preferably, the conversion of the maximum continuous month number of the past 12 month line usage >70% is a calculation of the maximum continuous month number usage of the past 12 month line usage >70% by using the original value, the conversion of the minimum cyclic credit line usage of the past 6 months is a calculation of the minimum cyclic credit line usage of the past 6 months by using the square root, the conversion of the current credit card remaining usable line by using the continuous method is a calculation of the current credit card remaining usable line by using the natural logarithm, the conversion of the past 3 month monthly credit present month account amount/month average asset scale value by using the cube root for the past 3 month monthly credit present month account amount/month average asset scale value, and the conversion of the current time point account by using the continuous method is a calculation of the current time point account use natural logarithm.
17. The method of claim 15, wherein,
The five feature-converted values of the maximum continuous month number of which the use rate of the credit of the past 12 months is more than 70%, the use rate of the credit of the cycle is the minimum value of the credit of the past 6 months, the current credit card remaining usable amount, the amount of money to be paid per month of the past 3 months monthly credit, the asset scale value of the deposit account balance of the current time point are substituted into a submodel constructed by adopting logistic regression based on sample retail credit prediction data to calculate the default probability of the sample to be predicted.
18. The method of claim 17, wherein,
The submodel is shown in the following formula 2:
Where k is the number of features of the model entered, preferably k is 5;
Alpha is the intercept term, preferably in the range (-0.27492, -0.66692), most preferably-0.47092;
β1 is the maximum continuous month number corresponding coefficient with the use rate of the forehead of the past 12 months >70%, and the preferred numerical range is (0.19904-0.03616), and the most preferred numerical range is 0.081443;
β2 is the minimum corresponding coefficient of the usage rate of the line of the past 6 months of the cyclic credit, and the preferred value range is (0.106, -0.09), and the most preferred value range is 0.008002;
beta 3 is the corresponding coefficient of the residual usable credit of the current credit card, and the preferred value range is (-0.49447, -0.76887), and most preferably-0.63167;
β4 is a corresponding coefficient of the amount of credit payoff per month of the last 3 months monthly/month average asset size value, preferably in the range of values (0.24108, -0.22932), most preferably 0.005877;
beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the preferred value range is (-0.6779, -0.9131), and the most preferred value range is-0.7955;
x1 is the original value of the maximum consecutive months of which the rate of use of the line of the past 12 months generated in the feature conversion step is > 70%;
x2 is the square root conversion value of the minimum value of the usage rate of the line of the past 6 months of the cyclic credit generated by the feature conversion step;
x3 is the natural logarithmic conversion value of the current credit card remaining usable credit generated in the feature conversion step;
x4 is the cube root conversion value of the asset scale of the amount to be refund/month average for the last 3 months monthly credits generated by the feature conversion step;
x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step.
19. The method of claim 18, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
20. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the minimum value of the usage rate of the credit for the line of the past 6 months, the usage rate of the credit for the line of the past 12 months of the cyclic credit is more than 10 percent, the usage rate of the credit for the line of the past 12 months is more than 70 percent, the asset size of the average asset of the past 6 months/the asset size of the average asset of the past 12 months, the usage rate of the line of the past 3 months is more than 90 percent, and one, two, three, four, five or six of the minimum balances of the time points of the deposit account of the past 3 months.
21. The method of claim 20, wherein the credit violation probability calculating step comprises:
Performing characteristic conversion on a minimum value of the utilization rate of the line of the past 6 months of the cyclic credit, a month number proportion of the utilization rate of the line of the past 12 months of the cyclic credit of >10%, a month number of the utilization rate of the line of the past 12 months of the credit of >70%, an average asset scale of the past 6 months/an average asset scale of the past 12 months, a month number of the utilization rate of the line of the past 3 months of >90%, and a minimum balance of the time points of the deposit account of the past 3 months,
Preferably, the minimum value of the use rate of the line of the past 6 months of the cyclic credit is converted in a continuous mode; converting the month number occupation ratio of the cyclic credit with the use ratio of more than 10% in the past 12 months by adopting a continuous mode; converting the number of months with the credit limit use rate of more than 70% in the past 12 months in a continuous mode; converting the asset size of the past 6 months/the asset size of the past 12 months in a continuous mode; the method comprises the steps of converting the number of months with the use rate of the line of 3 months more than 90% in the past by adopting a dummy feature mode; converting the minimum balance of the past 3 month deposit account time point in a continuous mode;
Further preferably, the minimum value of the usage rate of the line of the past 6 months of the cyclic credit is converted into a calculation mode of taking a cube root for the minimum value of the usage rate of the line of the past 6 months of the cyclic credit, the month number proportion of the line of the past 12 months of the cyclic credit with the usage rate of >10% is converted into a calculation mode of taking an original value for the month number of the line of the past 12 months of the cyclic credit with the usage rate of >10%, the method comprises the steps of converting the month number of which the credit use rate is more than 70% in the past 12 months into a calculation mode of using a raw value for the month number of which the credit use rate is more than 70% in the past 12 months, converting the past 6 month average asset scale/the past 12 month average asset scale into a calculation mode of using a cube root for the past 6 month average asset scale/the past 12 month average asset scale, and converting the minimum balance of the past 3 month deposit account time points into a calculation mode of using a natural logarithm for the past 3 month deposit account time points by using a continuous mode.
22. The method of claim 20, wherein,
Substituting the values converted by six characteristics of the minimum value of the utilization rate of the line of the circulation credit for the past 6 months, the utilization rate of the line of the circulation credit for the past 12 months >10% of the month number proportion, the utilization rate of the line of the credit for the past 12 months >70% of the month number proportion, the average asset scale of the past 6 months/the average asset scale of the past 12 months, the utilization rate of the line of the past 3 months >90% of the month number proportion and the minimum balance of the time point of the deposit account of the past 3 months into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
23. The method of claim 22, wherein,
The submodel is shown in equation 3 below:
Where k is the number of features of the model entered, preferably k is 6;
Alpha is the intercept term, preferably in the range (-0.23953, -0.27873), most preferably-0.25913;
β1 is the minimum value corresponding coefficient of the rate of use of the line of the last 6 months of the cyclic credit, and the preferred value range is (-0.4522, -0.4914), and most preferably-0.4718;
β2 is a month number corresponding coefficient with a cyclic credit usage rate of >10% for the past 12 months, and the preferred value range is (-0.22933, -0.26853), and most preferably-0.24893;
Beta 3 is a month number corresponding coefficient with a credit usage rate of 12 months in the past of >70%, and the preferred value range is (-0.51155, -0.58995), and most preferably-0.55075;
Beta 4 is the corresponding coefficient of the average asset size of the past 6 months/the average asset size of the past 12 months, and the preferable numerical range is (-0.19469, -0.27309), and the most preferable numerical range is-0.23389;
beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, and the preferable numerical range is (-0.22323, -0.26243), and the most preferable numerical range is-0.24283;
β6 is the minimum balance corresponding coefficient at the time of the deposit account of the past 3 months, and the preferred value range is (-0.80785, -0.84705), and most preferably-0.82745;
x1 is a cube root conversion value of the minimum value of the utilization rate of the line of the past 6 months of the cyclic credit generated in the characteristic conversion step;
x2 is the original value of the month number occupation ratio of the cyclic credit generated in the feature conversion step, wherein the use rate of the line of the past 12 months is more than 10%;
x3 is the original value of the number of months for which the credit generated in the feature conversion step has a usage rate of >70% for the past 12 months;
x4 is the cube root conversion value of the last 6 month average asset size/the last 12 month average asset size generated in the feature conversion step;
x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step;
x6 is the natural log-transformed value of the minimum balance of the past 3 month deposit account timepoints generated by the feature transformation step.
24. The method of claim 23, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
25. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: average asset size over the past 12 months, minimum balance of past 3 month financial account from present month, minimum balance of past 3 month monthly credit due to present month payoff/average asset size, minimum balance of past 6 month deposit account time point, month number of past 3 month line usage >90%, whether to send out wages, one, two, three, four, five, six and seven of past 6 month financial product balances/past 12 month financial product balances.
26. The method of claim 25, wherein the credit violation probability calculating step comprises:
The average asset scale of the past 12 months, the minimum balance of the past 3 months financial account, the minimum balance of the past 3 months monthly credit and the corresponding repayment/average asset scale of the month, the minimum balance of the past 6 months deposit account time point, the month number of the past 3 months use rate of >90 percent, whether to send out wages, the balance of the past 6 months financial products/the balance of the past 12 months financial products are subjected to characteristic conversion,
Preferably, the average asset size over the past 12 months is converted in a continuous manner; converting the minimum balance of the financial account of the past 3 months from the current month by adopting a continuous mode; converting the repayment/month-average asset scale of the monthly credit of the past 3 months by adopting a continuous mode; converting the minimum balance of the past 6 month deposit account time point in a continuous mode; the method comprises the steps of converting the number of months with the use rate of the line of 3 months more than 90% in the past by adopting a dummy feature mode; converting whether the generation wages adopt a dummy feature mode; converting the balance of the financial product of the past 6 months/the balance of the financial product of the past 12 months in a continuous mode;
further preferably, the method of calculating the natural logarithm of the average asset scale of the past 12 months is converted from the continuous method of calculating the natural logarithm of the average asset scale of the past 12 months, the method of calculating the natural logarithm of the minimum balance of the past 3 month financial account from the present month is converted from the minimum balance of the past 3 month financial account from the present month to the original value of the minimum balance of the past 3 month financial account, the method of calculating the square root of the average asset scale of the past 3 month monthly credit to the average asset scale of the past 3 month monthly credit to the average asset scale of the past 3 month, the method of calculating the minimum balance of the past 6 month deposit account from the minimum balance of the past 6 month deposit account, and the method of calculating the square root of the past 6 month financial balance/the past 12 month financial balance.
27. The method of claim 25, wherein,
Substituting values converted from seven characteristics of average asset size of the past 12 months, minimum balance of the past 3 month financial account, minimum balance of the past 3 month monthly credit due to repayment/average asset size of the month, minimum balance of the past 6 month deposit account time point, use rate of the past 3 month line >90%, whether to send wages, past 6 month financial product balance/past 12 month financial product balance into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
28. The method of claim 27, wherein,
The submodel is shown in equation 4 below:
where k is the number of features of the model entered, preferably k is 7;
alpha is the intercept term, preferably the range of values is (3.5122,3.4338), most preferably 3.473;
β1 is the average asset size corresponding coefficient over the past 12 months, with a preferred range of values (-0.1731, -0.2123), most preferably-0.1927;
β2 is the corresponding coefficient of the minimum balance of the financial account of the past 3 months to the number of months, and the preferred value range is (-0.21993, -0.29833), and the most preferred value range is-0.25913;
β3 is the corresponding coefficient of the size of the assets that should be repaid/month-average for the past 3 months monthly credit, the preferred range of values is (-0.22933, -0.26853), and most preferably-0.24893;
β4 is the minimum balance corresponding coefficient at the time of the deposit account of the past 6 months, and the preferred value range is (-0.53115, -0.57035), and most preferably-0.55075;
Beta 5 is the month number corresponding coefficient with the use rate of the previous 3 months limit >90%, and the preferable numerical range is (-0.27928, -0.35768), and the most preferable numerical range is-0.31848;
β6 is a payroll corresponding coefficient, preferably in the range (-0.21913, -0.25833), most preferably-0.23873;
beta 7 is the corresponding coefficient of the balance of the financial product in the past 6 months/the balance of the financial product in the past 12 months, and the preferable value range is (-0.55257, -0.63097), and the most preferable value range is-0.59177;
x1 is the natural logarithmic conversion value of the average asset size over the past 12 months generated by the feature conversion step;
x2 is the original value of the minimum balance of the financial account of the past 3 months generated in the feature conversion step from the number of months;
x3 is the cube root conversion value of the monthly payment/monthly average asset size of the past 3 months monthly credit generated in the feature conversion step;
x4 is a natural logarithmic conversion value of the minimum balance of the past 6 month deposit account time points generated in the feature conversion step;
x5 is a dummy feature conversion value of the number of months with a usage rate of >90% for the past 3 months generated in the feature conversion step;
x6 is a dummy feature conversion value of whether to replace wages generated in the feature conversion step;
x7 is the square root conversion value of the past 6 month financial product balance/the past 12 month financial product balance generated by the feature conversion step.
29. The method of claim 28, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
30. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: the cycle credit has a past 6 month line usage of >70% for one, two, three, four or five of the month number, past 3 month credit customer overdue balance average, past 6 month credit customer overdue balance percentage of the loan amount, credit customer current account age, credit past 3 month repayment amount maximum.
31. The method of claim 30, wherein the credit violation probability calculating step comprises:
performing characteristic conversion on the month number of which the cycle credit has the past 6 month limit use rate of more than 70 percent, the past 3 month credit customer overdue balance average value, the percentage of the past 6 month credit customer overdue balance accounting for the loan amount, the current account age of the credit customer and the maximum value of the past 3 month repayment amount of the credit,
Preferably, the month number of which the use rate of the line of the past 6 months of the cyclic credit is more than 70% is converted in a continuous mode; converting the overdue balance mean value of the past 3-month credit client in a continuous mode; converting the percentage of overdue balances of past 6-month credit clients in the loan amount by adopting a dummy feature mode; converting the current account age of the credit client in a WOE mode; converting the maximum value of the repayment amount of the past 3 months of credit in a continuous mode;
It is further preferable that the method of continuously converting the number of months of the past 6 month line usage rate >70% of the cyclic credit into the method of calculating the use of the original value of the number of months of the past 6 month line usage rate >70% of the cyclic credit, the method of continuously converting the average value of the past 3 month credit customer overdue balance into the method of calculating the use of the natural logarithm of the average value of the past 3 month credit customer overdue balance, and the method of continuously converting the maximum value of the 3 month refund amount of the past 3 month credit into the method of calculating the use of the square root of the maximum value of the 3 month refund amount of the credit.
32. The method of claim 30, wherein,
Substituting the numerical values converted by five characteristics of the past 6 month limit utilization rate of cyclic credit of 70 percent, the past 3 month credit customer overdue balance average value, the past 6 month credit customer overdue balance percentage of the loan amount, the current account age of the credit customer and the past 3 month repayment amount maximum value into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
33. The method of claim 32, wherein,
The submodel is shown in equation 5 below:
Where k is the number of features of the model entered, preferably k is 5;
alpha is the intercept term, preferably the range of values is (2.9266,2.7698), most preferably 2.8482;
β1 is a month number corresponding coefficient with a cyclic credit usage rate of >70% for the past 6 months, preferably in the range of (0.05201, -0.06559), most preferably-0.00679;
β2 is the average corresponding coefficient of the overdue balances of the credit customers over the past 3 months, and the preferred value range is (0.03734, -0.04106), and most preferably-0.00186;
β3 is a corresponding coefficient of the percentage of the past 6 months credit customer's overdue balance to the loan amount, preferably ranging in value from (-0.10043, -0.21803), optimally-0.15923;
Beta 4 is the current account age corresponding coefficient of the credit client, the preferred numerical range is (0.03485, -0.20035), and the most preferred numerical range is-0.08275;
beta 5 is the maximum value corresponding coefficient of the repayment amount of the past 3 months of credit, and the preferred value range is (-0.36818, -0.44658), and most preferably-0.40738;
x1 is the original value of the number of months that the cyclic credit generated in the feature conversion step has a usage rate of the line of the past 6 months of > 70%;
x2 is a natural logarithmic conversion value of the average value of the overdue balances of the past 3-month credit customers generated in the feature conversion step;
x3 is a dummy feature conversion value of the percentage of the overdue balance of the past 6 month credit customer to the loan amount generated in the feature conversion step;
x4 is the WOE conversion value of the current account age of the credit client generated in the feature conversion step;
x5 is the square root conversion value of the maximum value of the repayment amount of the credit for the past 3 months generated by the feature conversion step.
34. The method of claim 33, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
35. The method according to any one of claims 1 to 9, wherein,
The retail credit prediction data is selected from the group consisting of: one, two, three, four or five of the past 6 month gold account holding month number, the past 12 month investment financial account average balance, the past 12 month wage maximum, the past 3 month asset size/past 12 month asset size average, the current time point deposit account balance.
36. The method of claim 35, wherein the credit violation probability calculating step comprises:
performing characteristic conversion on the number of holding months of the past 6 month gold account, the average balance of the past 12 month investment financial account, the maximum wage value of the past 12 months, the average value of the past 3 month asset scale/the past 12 month asset scale and the balance of the deposit account at the current time point,
Preferably, the number of the past 6 months of the gold account holding month is converted in a continuous mode; converting the average balance of the investment financial account of the past 12 months in a continuous mode; converting the maximum wage value of the past 12 months in a dummy feature mode; converting the average value of the past 3 month asset scale/the past 12 month asset scale in a continuous mode; converting the balance of the deposit account at the current time point in a continuous mode;
Further preferably, the calculation method is a calculation method in which the past 6 months of the number of gold account holding months is converted into the original value used for the past 6 months of gold account holding months by a continuous method; the average balance of the investment financial account of the past 12 months is converted into a calculation mode of taking natural logarithm by adopting a continuous mode; converting the past 3 month asset scale/past 12 month asset scale average value into a calculation mode using a cube root for the past 3 month asset scale/past 12 month asset scale average value by adopting a continuous mode; the current time point deposit account balance is converted into a calculation mode of using natural logarithm for the current time point deposit account balance by adopting a continuous mode.
37. The method of claim 35, wherein,
Substituting the five feature-converted values of the past 6 month gold account holding month number, the past 12 month investment financial account average balance, the past 12 month wage maximum value, the past 3 month asset scale/past 12 month asset scale average value and the current time point deposit account balance into a submodel constructed by adopting logistic regression based on sample retail credit prediction data and credit violation probability to calculate the violation probability of the sample to be predicted.
38. The method of claim 37, wherein,
The submodel is shown in equation 6 below:
Where k is the number of features of the model entered, preferably k is 5;
Alpha is the intercept term, preferably in the range (-3.0574, -3.2926), most preferably-3.175;
β1 is the corresponding coefficient for the number of months the gold account has over the last 6 months, the preferred range of values is (-0.76852, -0.96452), and most preferably-0.86652;
β2 is the average balance corresponding coefficient of the investment financial account of the past 12 months, the preferred value range is (-0.37245, -0.64685), and the most preferred value range is-0.50965;
β3 is the maximum payroll corresponding coefficient for the past 12 months, with a preferred range of values (-0.81714, -1.28754), and most preferably-1.05234;
beta 4 is the average corresponding coefficient of the past 3 month asset scale/the past 12 month asset scale, the preferred value range is (-0.0053, -0.2405), and most preferably-0.1229;
beta 5 is the corresponding coefficient of the deposit account balance at the current time point, the preferred value range is (-0.92484, -1.04244), and the most preferred value range is-0.98364;
x1 is the original value of the number of months held by the gold account for the last 6 months generated in the feature conversion step;
x2 is a natural logarithm conversion value of the average balance of the investment financial account of the past 12 months generated in the characteristic conversion step;
x3 is a dummy feature conversion value of the maximum payroll value of the past 12 months generated in the feature conversion step;
x4 is the cube root transformed value of the past 3 month asset size/past 12 month asset size average generated by the feature transformation step;
x5 is the natural log-transformed value of the current time point deposit account balance generated by the feature transformation step.
39. The method of claim 38, wherein,
After calculating the credit breach probability, calculating a credit score for the sample to be predicted is calculating a credit score for characterizing the borrower using the formula:
Wherein P is the default probability (P) of the borrower generated in the credit violation probability calculation module, and A is 443.9036; b is-72.1348, and the round function rounds up the rounding value after calculating the score; and finally, setting the score of more than 1000 as 1000 and the score of less than 0 as 0.
40. An apparatus for calculating retail credit risk, comprising:
The data acquisition module is used for acquiring retail credit prediction data of a sample to be predicted;
A module for classifying the sample to be predicted, for classifying the sample to be predicted based on a decision tree method to determine a sub-model for calculating a credit violation probability;
A credit breach probability calculation module for substituting retail credit prediction data into a credit breach probability submodel to calculate a credit breach probability for the sample to be predicted.
41. The apparatus of claim 40, wherein the apparatus performs the steps of the method of calculating retail credit risk of any of claims 1 to 39.
42. A system for calculating retail credit risk, the system for calculating retail credit risk comprising: a memory, a processor and a program stored on the memory and executable on the processor for the method of calculating retail credit risk, the program for calculating retail credit risk when executed by the processor implementing the steps of the method of calculating retail credit risk as recited in any one of claims 1 to 39.
43. A computer storage medium having stored thereon a program for a method of calculating a risk of retail credit, the program for a method of calculating a risk of retail credit, when executed by a processor, implementing the steps of the method of calculating a risk of retail credit as recited in any one of claims 1 to 39.
44. A method of constructing a retail credit risk prediction model for a consumer credit business, comprising:
a data acquisition step of acquiring raw retail credit prediction data of a sample used for constructing a model;
a data deriving step of processing derived retail credit prediction data based on the original retail credit prediction data;
A feature preliminary screening step of preliminarily screening all categories, i.e., all features, including the original retail credit prediction data and the derived retail credit prediction data to obtain preliminarily screened features;
a preliminary screening data conversion step of judging a conversion mode of the preliminarily screened features to confirm that one of a WOE conversion mode, a dummy feature conversion mode and a continuous conversion mode is adopted to perform feature conversion, and performing feature conversion by adopting a judged optimal mode for each preliminarily screened feature;
A feature fine screening step, namely performing deep screening on the features subjected to the primary screening after feature conversion to obtain the features subjected to the fine screening;
A credit violation probability modeling step, namely selecting a logistic regression mode for the combination of the characteristics after fine screening and the probability relation between the credit violations to carry out model construction, and confirming a mode for calculating the credit violation probability;
Wherein the data sample obtained in the data collection step is a customer sample that has used the consumer credit service.
45. The method of claim 44, wherein,
In the data acquisition step, the raw retail credit prediction data of the sample for constructing the model obtained includes:
credit card based data, which is based on all available data in the sample user's credit card creation process and use process,
Personal loan class base data, which is all available data based on the sample user's loan application case and usage behaviors,
Customer base information class base data, which is data based on the properties of the sample user itself, but not directly related to behavior at the financial institution,
Personal financial property type base data, which is sample user data of all other financial properties and financial transaction types not related to credit cards and loans at a financial institution.
46. The method of claim 44, wherein,
In the data deriving step, processing derived retail credit prediction data based on the original retail credit prediction data refers to data obtained by processing the collected original retail credit prediction data based on a time dimension, a space dimension, a frequency dimension, and a statistical information dimension;
preferably, the derived retail credit prediction data includes, but is not limited to:
derived retail credit prediction data processed based on sample relationship lengths,
Derived retail credit forecast data processed based on time interval class variables,
Derived retail credit forecast data processed based on the degree of sample behavioral frequency,
Derived retail credit prediction data processed based on the sample current point in time conditions,
Derived retail credit forecast data processed based on sample duration behavior,
Derived retail credit prediction data processed from the sample data based on the statistical information dimension.
47. The method of any one of claims 44-46, wherein,
The feature primary screening step comprises the following steps:
A first preliminary screening step of screening the features based on the data missing for each feature of the sample used to construct the model,
A second preliminary screening step of screening the features based on the condition that the single value of a certain feature sample is too high,
A third preliminary screening step, namely calculating the information IV value of each feature to carry out preliminary screening on the features;
The order of the first preliminary screening step, the second preliminary screening step and the third preliminary screening step may be any order,
A fourth preliminary screening step, wherein the features subjected to the first to third preliminary screening are subjected to preliminary screening by adopting a gradual discrimination algorithm;
And a fifth preliminary screening step, wherein the features after the fourth preliminary screening step are subjected to preliminary screening based on the coincidence condition of the risk characteristics of the features and the actual and real results of the samples for model construction.
48. The method of any one of claims 44-47, further comprising:
A sample selection step for screening all users and acquiring a sample using the consumer credit service for model construction prior to the data acquisition step,
Preferably, the sample selection step includes classifying all users of the sample based on a decision tree, the classification basis including but not limited to:
Whether a user is a customer who has applied for registering credit services at a financial institution;
whether a certain user is a customer who has a credit overdue;
whether the current monthly credit utilization rate of a certain user is more than or equal to a percent;
whether the current monthly asset size of a certain user is larger than or equal to b.
49. The method of any one of claims 44 to 48, wherein in the preliminary screening data conversion step, the conversion of the preliminary screened features is determined based on the concentration and data type of the preliminary screened features.
50. The method of claim 49, wherein,
The primary screening data conversion step based on the judgment of the concentration and the data type comprises the following steps:
classifying the data type for each feature classifies each feature into character-type variables and numerical-type variables,
The character type variable is subjected to preliminary screening data conversion by adopting a dummy feature conversion mode,
The process of further classifying the numerical variables comprises the following sub-steps:
If the numerical variable has less than n values, performing primary screening data conversion by adopting a WOE conversion mode,
If the value of the numerical variable is more than n, further judging that if the numerical variable is converted into continuous variable with more values and the concentration degree of single value is more than m%, adopting a WOE conversion mode, if the concentration degree of single value is less than or equal to m%, adopting a continuous conversion mode,
Preferably, n and m are both positive integers, where n=5 to 10 and m=90 to 99.
51. The method of claim 50, further comprising:
An optimal conversion method is selected for confirming the feature adopting the continuous conversion mode based on the correlation of the feature with credit violations under different continuous conversion modes to perform continuous feature conversion of the feature,
The continuous feature transformation is preferably performed by directly selecting the original value, calculating the square of the original data, calculating the square root of the original data, calculating the cube root of the original data, or calculating the natural logarithm of the original data.
52. The method of any one of claims 44 to 51, wherein the feature sizing step comprises:
A first fine screening step, based on a stepwise regression algorithm, of screening features based on the significance of the features by the F test and the T test,
A second fine screening step of calculating a variance expansion factor based on each feature and eliminating features having a higher variance expansion factor to screen the features,
And a third fine screening step of analyzing whether the characteristic coefficient accords with the trend of the prediction result aiming at credit violations or not based on the logistic regression pair after the first fine screening step and the second fine screening step so as to further perform characteristic screening.
53. The method of any one of claims 44 to 52, wherein the credit breach probability modeling step substitutes the features filtered by the feature fine screening step into a Sigmoid function for logistic regression to calculate a model of credit breach probability.
54. An apparatus for constructing a retail credit risk prediction model for a consumer credit service, the apparatus comprising:
A data acquisition module for acquiring raw retail credit prediction data for a sample used to build a model;
a data derivation module for processing derived retail credit prediction data based on the original retail credit prediction data;
the feature primary screening module is used for carrying out primary screening on all categories comprising original retail credit prediction data and derived retail credit prediction data, namely all features, so as to obtain features after primary screening;
The primary screening data conversion module is used for judging a conversion mode of the features after primary screening so as to confirm that one of a WOE conversion mode, a dummy feature conversion mode and a continuous conversion mode is adopted for carrying out feature conversion, and carrying out feature conversion by adopting an optimal mode for judging each feature after primary screening;
The feature fine screening module is used for carrying out deep screening on the features subjected to the primary screening after feature conversion so as to obtain the features subjected to the fine screening;
A credit violation probability modeling module for model construction by selecting a logistic regression mode for the probability relation between the feature combination after fine screening and the credit violation, and confirming the mode for calculating the credit violation probability,
The data sample acquired by the data acquisition module is a customer sample using the consumed credit business.
55. The apparatus of claim 54, wherein the apparatus performs the steps of the method of constructing a retail credit risk prediction model of any one of claims 44 to 53.
56. A system for constructing a retail credit risk prediction model for a consumer credit service, the system comprising: a memory, a processor and a program stored on the memory and executable on the processor to construct a retail credit risk prediction model method, the program to construct a retail credit risk prediction model method when executed by the processor implementing the steps of constructing a retail credit risk prediction model method as claimed in any one of claims 44 to 53.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211324673.3A CN117994016A (en) | 2022-10-27 | 2022-10-27 | Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211324673.3A CN117994016A (en) | 2022-10-27 | 2022-10-27 | Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117994016A true CN117994016A (en) | 2024-05-07 |
Family
ID=90899806
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211324673.3A Pending CN117994016A (en) | 2022-10-27 | 2022-10-27 | Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117994016A (en) |
-
2022
- 2022-10-27 CN CN202211324673.3A patent/CN117994016A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109035003A (en) | Anti- fraud model modelling approach and anti-fraud monitoring method based on machine learning | |
Koh et al. | A two-step method to construct credit scoring models with data mining techniques | |
CN110807700A (en) | Unsupervised fusion model personal credit scoring method based on government data | |
CN104321794A (en) | A system and method using multi-dimensional rating to determine an entity's future commercial viability | |
CN112700324A (en) | User loan default prediction method based on combination of Catboost and restricted Boltzmann machine | |
CN113095927A (en) | Method and device for identifying suspicious transactions of anti-money laundering | |
CN107633455A (en) | Credit estimation method and device based on data model | |
CN111738819A (en) | Method, device and equipment for screening characterization data | |
CN111709826A (en) | Target information determination method and device | |
CN114078050A (en) | Loan overdue prediction method and device, electronic equipment and computer readable medium | |
CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
CN112232950A (en) | Loan risk assessment method and device, equipment and computer-readable storage medium | |
CN117114812A (en) | Financial product recommendation method and device for enterprises | |
CN113421154B (en) | Credit risk assessment method and system based on control chart | |
CN115809930A (en) | Anti-fraud analysis method, device, equipment and medium based on data fusion matching | |
CN114626940A (en) | Data analysis method and device and electronic equipment | |
CN115099933A (en) | Service budget method, device and equipment | |
CN114612239A (en) | Stock public opinion monitoring and wind control system based on algorithm, big data and artificial intelligence | |
CN117252677A (en) | Credit line determination method and device, electronic equipment and storage medium | |
Lee et al. | Application of machine learning in credit risk scorecard | |
Pradnyana et al. | Loan Default Prediction in Microfinance Group Lending with Machine Learning | |
CN113822751A (en) | Online loan risk prediction method | |
CN117994016A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebeta model | |
CN118333738A (en) | Method for constructing retail credit risk prediction model and credit card service Scorealpha model | |
CN118071482A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebetad model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |